irHE AUTONOMIC NERVOUS SYSTEM - Internet Archive

^^—irHE AUTONOMICNERVOUS SYSTEM

i

PART I.

J. N. LANGLEY. F.R.S.

e'

CAMBRIDGE

:

W. HEFFER e? SONS LTD.

1921.

—^^

QJarnell ItttoerHltg ICtbrarg

Stifatu, New fork

BOUGHT WITH THE INCOME OF THE

SAGE ENDOWMENT FUNDTHE GIFT OF

HENRY W. SAGE1691

QP 368.L28"'"""'"""•* '"'"'^

3 1924 024 561 908

|| Cornell University

J Library

The original of tiiis book is in

tine Cornell University Library.

There are no known copyright restrictions in

the United States on the use of the text.

http://www.archive.org/details/cu31924024561908

THEAUTONOMIC NERVOUS SYSTEM

LONDON AGENTS

SIMPKIN, MARSHALL, HAMILTON, KENT & Co. Ltd.

THE AUTONOMICNERVOUS SYSTEM

BY

J. N. LANGLEY,Sc.D., Hon. LL.D., Hon. M.D., F.R.S.

Professor of Physiology in the University of Cambridge

PART I.

CAMBRIDGE

W. HEFFER & SONS LTD.

1921

w

^5\5•]4-2.

V\0 U\cna. Wolitlied

A.^l.

-"r.

PAGE

Contents

1. The Divisions of the Autonomic System.Nomenclature ----- j

y.'Oies.—Nomenclature. Voluntary Production ofAutonomic Effects, p. 12.

2. General Plan of Origin and of Peri-pheral Distribution - - - - 15

3. The Nerve Fibres of the AutonomicSystem ------ 22

4. The Specific Action of Drugs on the Sym-pathetic AND Parasympathetic Systems 28

a. Adrenaline and Sympathetic Effects 29Notes.—Fall of Blood Pressure and Dilatation of

Blood Vessels; action on Cerebral Vessels,Veins, Capillaries, p. 31. Action on Muscle,

P- 33-

b. Pilocarpine and ParasympatheticEffects ----- 35

Notes.—Action on Uterus and Blood Vessels,p. 36.

c. Reversal of Effect of SympatheticAND Parasympathetic Drugs - - 37

d. Theories on the Relation of Drugsto Nerve Systems - - - - 39

Note.—Localisation op Action of Drugs on StriatedMuscle, p. 47.

e. Classification of Sympathetic andParasympathetic Nerves accordingto Pharmacological Results - - 50

Notes.—Historical. Parasympathotonia Sympatho-TONIA. PerIT.ERMINAL NETWORK, p. 53.

5. The Tissues Innervated - - - 57

a. Pigment Cells- - - - - 59

b. Sympathetic Nerves and the Inner-

vation of Capillaries - - - 64Notes.—Capillary Contractility, p. 67.

c. Sympathetic Nerves and the Inner-

vation of Striated Muscle - ' - 69

Some Abbreviations in Reference

to Papers

A.

1. The Divisions of the Autonomic

System. Nomenclature.

The autonomic nervous system consists of the nerve

cells and nerve fibres, by means of which efferent

impulses pass to tissues other than multi-nuclear

striated muscle.

The progress of knowledge of this system has been

continuous for at least the last 250 years, though at

times it has progressed but slowly. As knowledge

increased, new points of view presented themselves,

and the terms used to express the general conceptions

naturally varied. I give a short historical account of

these terms, omitting, as far as practicable, reference

to the actual state of knowledge at the successive

periods.

Voluntary and involuntary. In the early part

of the 1 8th century the movements of the various

parts of the body were commonly divided into three

classes, viz. (i) voluntary movements, (2) involuntary

movements which could also be produced bjr the will,

such as the movements of the respiratory muscles in

sleep and instinctive movements, (3) involuntary

movements over which the will had little or no control,

such as the movements of the heart and intestines

;

this form of involuntary movement was called vital

or natural movement.

In later times, in such observations as were mainly

confined to the vital movements, the nerves supposed

to be concerned with them were not infrequently

2 AUTONOMIC NERVOUS SYSTEM

spoken of as involuntary nerves, but in a general

classification, a division into voluntary and in-

voluntary nerves was rarely adopted. The use of

"involuntary" in more or less of the sense in which

we have used the term autonomic occurs however

throughout the 19th century, and was so used by

Gaskell in his last work in 19 14.

The sympathetic nerves. A different termin-

ology began with Winslow (1732). The nerve called

the intercostal nerve was known to be the chief nerve

to send fibres to the intestines, and it was known that

it also sent fibres to the heart. In consequence of its

numerous connexions it was supposed to be the

means by which one part of the body affected another

part, or, as it was then stated, by which the sympathies

of the body were brought about. Winslow, in view

of this, called the nerve the great sympathetic. Twoother nerves Winslow considered had a similar

function, viz., the portio dura of the seventh nerve,

which he called the small sympathetic, and the vagus,^

which he called the medium sympathetic.

The terms small sympathetic and medium sympa-

thetic were rarely used. In France "great sympa-

thetic" remained as the designation of the intercostal

nerve; in other countries, and sometimes in France,

"great" was dropped, and the intercostal nerve

became the sympathetic nerve, and with its branches,

the sympathetic nervous system. The meaning of

sympathetic nervous system has from time to time

been extended to include more or less of the rest of the

autonomic system.

Ganglionic nerves and ganglionic nervous

SYSTEM. A third terminology arose from the investi-

gation of the ganglia on the course of the nerves.

NOMENCLATURE 3

Johnstone (1764) put forward the theory that the

gangha served to convert voluntary into involuntary

movements. Thus the nerves governing involuntary

movements could naturally be spoken of as ganglionic

nerves. The ganglionic nerves were not confined to

the nerves governing vital movements, but included

the nerves governing the involuntary movements of

muscles which could also be put in action by the will,

and thus the ganglia on the course of the cranial and

spinal nerves were included in the ganglionic nervous

system. In the course of time most of the cells of

these ganglia were recognised as belonging to afferent

nerves, and were excluded from the ganglionic

system, but the term was rarely, if ever, used to

exclude them all.

Nervous system of animal and of organic life.

Additional terms were adopted by Bichat (i 800-1).

On the lines of suggestions made b}' some earlier

writers, he divided the life of the organism into the

animal life (vie de relation) and the vegetative or

organic life (vie de nutrition). There were thus

vegetative and animal nerves. The ganglia of the

great sympathetic and some of those of the cranial

nerves, he said—-with various reservations and ob-

scurities—^were the nervous organs of organic life.

The theory was much discussed, but was either

rejected or modified by most of those who submitted

it to the test of experiment. It increased for a time

the use of the terms ganglionic nerves and ganglionic

nervous system, and not infrequently led to the use of

the terms organic nerves and organic nervous system.

The theory and the term organic nearly died out about

the middle of the 19th century. "Vegetative" has

recentlv been revived as a substitute for " autonomic."


Cerebro-spinal and sympathetic nerves. In

the first half of the 19th century, concurrently with

the use of the terms mentioned above, it was not un-

common to divide the nerves into cerebro-spinal

nerves and their ganglia on the one hand, and sympa-

thetic nerves and their ganglia on the other, and this

became customary from about the middle of the 19th

century until 1 884-1 886. Both systems supplied

"involuntary" organs and tissues. The resemblance

of certain branches of the cranial nerves and of the

visceral branches of the sacral nerves to the splanchnic

nerves were from time to time pointed out, but there

was no general grouping of the nerves supplying

unstriated muscle and glands into a single system.

A partial return to an earlier nomenclature was

advocated by Dastre and Morat in 1 884. It was based

on their observations on vaso-motor nerves. Theydivided motor nerves into non-ganglionated nerves

subserving the "vie de relation" of Bichat, and

ganglionic nerves subserving Bichat 's "vie de nutri-

tion," excluding, however, nutrition and assimilation.

Under the general head of the sympathetic or gangli-

onic system they included certain unnamed branches

of the 5th, 7th, 9th and loth cranial nerves, so that

the sympathetic system was defined in much the

same way as it had been defined by Winslow.

Visceral nerves. Ganglionated splanchnic

nerves. Gaskell (i 886-1 889) considered the question

of the classification of nerves chiefly from a morpho-

logical standpoint. He said that certain only of the

cerebro-spinal nerves had a "visceral" branch in the

sense of the morphologist. These branches he placed

together as visceral nerves. They arose from homo-

logous columns of cells in the central nervous system.

NOMENCLATURE 5

broken by the development of the nerves to the hmbs;

the vagus was homologous with the hypogastric nerve,

and the nervus erigens with the splanchnic nerve.

Thus the visceral nerves left the central nervous

system in three outflows—-the cranial, the thoracic, and

the sacral. This was the first definite statement that

the sympathetic did not receive branches from each

spinal nerve. From another point of view he divided

the nerves into somatic and splanchnic, each having

a ganglionated and a non-ganglionated part. The

visceral nerves formed the ganglionated splanchnic

part. The ganglia on the course of the visceral

nerves Gaskell classified as (i) proximal or vertebral

ganglia, i.e. the ganglia of the sympathetic chain from

the first thoracic ganglion downwards; and (2) distal

ganglia, consisting of (a) pre-vertebral ganglia, which

included the superior cervical, the semilunar and the

inferior mesenteric ganglia; and (h) terminal ganglia,

which were scattered in or near the tissues. Gaskell

considered (1886 p. 30) that the blood vessels of the

mylohyoid muscle of the frog, the vessels of the

limbs of mammals, and possibly other vessels, in

addition to receiving nerve fibres from visceral nerves,

received efferent vaso-dilator fibres which ran direct

outward in the roots of the nerves.

Autonomic. The observations made by Dickinson

and myself ( 1 889) on the action of nicotine gave a new

method of investigating the connexion of nerve fibres

with peripheral nerve cells. In describing the results

of such investigations and of others suggested by it,

I felt the need of a new term for the system of nerves

I was dealing with. All the old terms had been used

with different connotations at different times, and to

use anv of them with another additional connotation


was but to add to the inherent difficulty of under-

standing the point of view of earher writers. I called

the system the autonomic nerve system (1898). It

was a "local" autonomy that I had in mind. Theword "autonomic" does suggest a much greater

degree of independence of the central nervous system

than in fact exists, except perhaps in that part which

is in the walls of the alimentary canal. But it is, I

think, more important that new words should be used

for new ideas than that the words should be accurately

descriptive. In any case the old terms have no

advantage as descriptive terms.

"Vegetative" implies a contrast between plant and

animal life, and to speak of a "vegetative nervous

system" is to obscure one of the fundamental differ-

ences between plants and animals, viz., the absence of

nerves in the one, and their presence in the other.

Overlooking this we might still use "vegetative" for

the nervous system in animals governing those

processes in them which occur also in plants. But

assimilation and metabolism are conspicuous plant

processes, and control of these processes in animals

is not a special function of the "vegetative" nervous

system.

The fundamental drawback to the use of'

' involun-

tary" as a designation for the nervous system we are

considering is that it makes subjective sensations the

criterion of classification. It is inappropriate in a

science based on objective observation. The term

also implies that all other movements are "voluntary,"

whereas, in fact, all "voluntary" muscles are set in

action involuntarily, and some more often than

voluntarily. Further, there are a few cases of muscles

which histologically belong to the class of " voluntary"

NOMENCLATURE 7

muscle, but which are apparently not controlled bythe will. There is no evidence that the striated muscle

of the oesophagus can be contracted voluntarily ; so

far as it is known it is only set in action by initiating

swallowing, i.e. it is set in action in exactly the same

way as the "involuntary" unstriated muscle of the

oesophagus. It is difficult to believe that a frog can

voluntarily control its l3niiph hearts, or that a bird

can contract its iris at will. Lastly, it is untrue that

the " involuntary " actions are out of all control of the

will. What is true is that the nervous mechanism set

in action is different from that set in action when, for

example, a limb is voluntarily moved. The will

exercises more or less control over unstriated muscles

and glands by recalling emotions and sensations. Andapart from any special emotion, some people can by

effort of will cause contraction in the involuntary

unstriated muscle of the skin, and others can cause

acceleration of the heart. Some physiologists have

claimed that they can contract the bladder at will,

and cases are recorded of voluntary inhibition of the

heart, and contraction of the pupil. It can hardly be

doubted that many more instances would be found if

sufficient attention were directed to the subject.

The term " ganglionic " is misleading since there are

ganglia on the course of all the spinal nerves and on

most of the cranial nerves, and since historically the

"ganglionic system" was long used in its natural

signification to include these ganglia.

Sympathetic nerves have no special relation to

sympathies. But the chief objection to calling the

whole autonomic system sympathetic is that it con-

fuses instead of simplifying nomenclature. Nearly all

observers have used it for what was formerly called


the intercostal nerve, and rarely if ever has it been

taken to include the nervus erigens.

At the time I introduced the term " autonomic

"

there were two points of view from which the innerva-

tion of unstriated muscle and glands were regarded.

From one point of view these tissues were supplied

with nerve fibres partly by cerebro-spinal nerves and

partly by sympathetic nerves. From the other (that

of Gaskell) they were supplied by one system, which

anatomically was separated into three parts by the

development of the nerves to the limbs. Neither of

these seemed to me properly to express the conditions.

The facts that the sympathetic innervated the whole

body, whilst the cranial and sacral outflows innervated

parts only, and that the sympathetic had, in general,

opposite functional effects from those of the other

autonomic nerves, indicated that the sympathetic

was a system distinct from the rest. The part of the

cranial outflow suppl3dng the eye seemed clearly to be

distinct from the rest of the cranial outflow. Further,

the bulbar part of the cranial outflow and the sacral

outflow seemed equally clearly to form one system

innervating the alimentary canal and parts develop-

mentally connected with it. I divided (1898) the

autonomic system into tectal, bulbo-sacral, and

sympathetic systems, and considered that each had a

different developmental history.

The theory that there is some fundamental differ-

ence between the sympathetic and the rest of the

autonomic system was much strengthened by the

discovery that the effects produced by adrenaline

were apparently confined to effects caused by stimu-

lating sympathetic nerves (cp. p 29). Since other

drugs caused effects more or less confined to those

NOMENCLATURE g

produced by stimulating tectal and bulbo-sacral

nerves, it was convenient to have a common name for

these nerves, and I placed them together as the

parasympathetic system (1905). The pharmaco-

logical relations of the nerve systems and some

variations in the meaning attached to parasympa-

thetic I discuss later.

I pointed out (1900) that we should expect the cells

of Auerbach's and Meissner's plexuses to be on the

course of the bulbar and sacral nerves, but as there

was no clear proof of their central connexion, and as

their obvious histological characters differed from

those of any other peripheral nerve cell, I placed them

in a class by themselves as the enteric nervous system.

This classification is, I think, still advisable, for the

central connexion of the enteric nerve cells is still

uncertain, and evidence has been obtained that they

have automatic and reflex functions which other

peripheral nerve cells do not possess. Gaskell, in his

postumous work (191 6), advocated on developmental

grounds the possible theory that the enteric nerve

system is part of the bulbo-sacral. He did not, how-

ever, settle any of the doubtful points.

Most previous observers who had discussed the

question considered that the posterior root ganglia

contained some cells of the type of the sympathetic,

and that the sympathetic contained cells of the type

of the posterior root ganglion, some left the question

open. My observations led me to believe that there

was no such admixture of the different type of cells.

On this theory all autonomic nerves were motor, and

the afferent nerves accompanying it were not as a

whole distinguishable from the other afferent nerves.

How far this is true will be discussed later.


It was convenient to have some term for the

general body nerves which did not belong to the

autonomic systeni, and I used the term "somatic"

since it expressed fairly well the character of the

nerves.

The statement of Strieker and others that efferent

vaso-dilator fibres ran direct to the limbs by way of

the posterior roots was confirmed by Bayliss (1900).

I had come to the conclusion that the vaso-dilator

effect was produced by afferent fibres, and was due to

setting free of metabolites. This theory I communi-

cated to Bayliss, through Prof. Starling, and suggested

that he should try the effect of stimulating the

posterior roots after time had been allowed for

degeneration. He found that the vaso-dilator effect

could still be obtained, but attributed it (1901) to a

direct action in the blood vessels. The nature and

extent of this "antidromic" action I shall discuss

later. For the present it is, I think, best classed as a

property of afferent somatic nerves.

The peripheral nerves, then, excluding the olfactory,

optic and auditory, may be classified as follows:

NervesI

I . I

Somatic AutonomicI

Afferent Nerves| | |

(Dromic and Anti- Sympathetic Parasympathetic Entericdromic action)

Efferent to Striated

Muscle

(Plexuses of Auerbachand Meissner)

Ocular Bulbo-sacral(oro-anal)

I I

Thoracico- Tectal Bulbar Sacrallumbar

Some writers have restricted the term autonomic to

NOMENCLATURE n

the parasympathetic system. There are other points

in nomenclature which may be mentioned here.

The terms vertebral and prevertebral ganglia do not

seem to me satisfactory. The cervical ganglia are

prevertebral in mammals and vertebral in other

vertebrates, so that the terms suggest their homology

with the ganglia of the abdominal viscera in the

mammal, and with the ganglia of the lower part of the

chain in other vertebrates.

It is frequently necessary to describe the state of a

tissue after the postganglionic fibres running to it have

been cut, or the ganglion supplying it with nerve

fibres has been excised, and time has been allowed for

degeneration of the cut fibres. The operation is

conveniently described as a degenerative section.

I use the terms paralysis of a nerve, as it is usually

used, to signify that stimulation of the nerve has no

effect, and not that the excitability or conductivity

of the nerve has been abolished.

The terminology expressing the point of action of a

chemical substance varies with different writers. The

following is, I think, the most convenient. The point

of action is

:

the end apparatus, when it is not desired to specify

anything further than that the action is at the

periphery and can be set in action by the nerve,

the nerve ending, in its usual significance of an action

on the terminal branches of the nerve,

the myo-neural, or neuro-cellular junction when it is

held that there is continuity between nerve and cell,

and that the action is on a partly nervous, partly

muscle substance at the junction,

the neural region when it is held that the action is on


the cell and more or less confined to the region of the

nerve ending,

the synapse, when it is held that there is no continuity

of nerve and cell and that the action is a physical one

in the neural region,

the receptive substance when it is held that the action

is brought about by a chemical combination with a

special cell constituent.

REFERENCES AND NOTES.

Winslow. Exposition Anat. de la Structure du Corps

Humain. (Paris.) 1732.

Johnstone. Phil. Trans. London, 54, p. 177, 1764. Essay

on the use of the Ganglions of the Nerves. (Shrewsbury.)

1771.

Bichat. Rech. Physiol, sur la vie et la Mort. (Paris.) 1800.

Anat. gen. and Anat. descriptive. (Paris.) iSoi.

Pastre and Morat. Rech. Exp. sur le Syst. Vaso-Moteur.

(Paris.) p. 330, 1884.

Gaskell. J. Physiol. 7, p. i, 1886; 10, p. 153, 1889.

Langley. Text book of Physiol., ed. by Schafer, 2, p. 694,

1900. Brain, 26, p. i, 1903. Ergeb. d. Physiol.,

2. Abt. 2, p. 819, 1903.

Bayliss. J. Physiol. 25. Proc. Physiol. Soc, p. xiii., 1900.

J. Physiol. 26, p. 173, 1901.

Gaskell. The Involuntary Nervous System, 1914. Pub-lished postumously. (London.) 1916.

The terms I have used were first employed as follows :

—

Pre- and postganglionic fibres, Proc. Roy. Soc. 52. p. 548, 1893.

Autonomic. J. Physiol. 23, p. 241, 1898. Axon reflex. Soc. de

Biol. T. JubUaire, 1899, p. 220. Bulbar and sacral nerves as

one system, Brit. Assoc. Adv. Science, Pres. Address, Physiol.

Sect., 1899; Mid-brain autonomic. Brain, 1903. Para-

sympathetic, Address to Students of Med. and Nat. PhU. in

Amsterdam, 1905; also J. Physiol. 33, p. 403, 1905; 43.

p. 173, 1911.

Gaskell (1886-1889) divided the nerves, and the tissues

innervated by them, into two groups, splanchnic and somatic

NOMENCLATURE 13

which corresponded more or less closely with those described

as arising in the embryo from a branchiomeric or dorsal

segmentation and from a mesomeric or ventral segmentationrespectively. This nomenclature made the masticatory,

respiratory and facial muscles, splanchnic; only the limb andmain trunk muscles being somatic, and it did not seem to meto be in harmony with the connotation of the terms in

Physiology. In consequence, and because branchiomeric

could be used when necessary for the "splanchnic" muscles, I

employed somatic for all motor nerves which were not auto-

nomic. (Brit. Assoc. Address, 1899) and later with Andersonto include also afferent fibres arising from spinal and the

similar cranial ganglia. (J. Physiol. 31. p. 380, 1904.) Therelation of the tissues to different segmentations was followed

up by GaskeU in connexion with his theory of the origin of

vertebrates. In his work published postumously (1916) his

views on the phylogenetic history of the autonomic system

are given. He considered that vertebrates arose from anancestor having two serially arranged segmentations, one

connected with the appendages—the splanchnic segmentation,

and the other with the body—the somatic segmentation.

The splanchnic segmentation gave rise to the muscles of the

gut with the exception of the sphincters, and of parts develop-

mentally connected with it. The somatic segmentation gave rise

to the other unstriated muscles. Thus some striated muscles

were splanchnic, and some unstriated muscles were somatic.

Both autonomic and somatic systems consist of central and

peripheral parts. The central connexions of the autonomic

system are imperfectly known, so that description of this

system is mainlj' confined to a description of the preganglionic

and postganglionic nerve connexions. A limitation of

"autonomic" to the peripheral nerve cells was proposed byHeubner (Zntrlb. Physiol. 1912, p. 1180). Similarly Gaskell

(1914) defined the involuntary nerve system as being limited

to the peripheral neurons; the preganglionic fibres he called

"connector." The limitation involves considerable diffi-

culties, and Gaskell not infrequently includes the connector

fibres in the involuntary system. Moreover, all nerve fibres

are connector, and to distinguish the different connector fibres

means much periphrasis. Lastly, it may be mentioned that

Gaskell used "enteral" as synonymous with bulbo-sacral.

Voluntary production of autonomic effects. Goose skin and

erection of hairs. Chalmers (J. of Physiol. 31, 1904, Proc.


Physiol. Soc, p. Ix.). The action was not restricted to one

region, but was produced more strongly in any desired region.

Maxwell (Amer. J. Physiol. 7, p. 369, 1902), Kcenigsfeld and

Zierl (Deut. A. f. klin. Med. 106, p. 442, 1912). Goose skin

was produced on one side of the body on imagining this to be

cold and the other side to be hot. Acceleration of heart, see

Papers quoted by White (Heart, 6, p. 175, 1917) ; in White's

case voluntary acceleration was also obtained after atropine.

King (Bull. J. H. Hosp. 31, p. 303, 1920). For some observa-

tions on the pupil cp. Lewandowsky (Zts. f . d. ges. Neurol. 14,

p. 282, 1913).

2. General Plan of Origin and of

Peripheral Distribution.

At this stage it is only necessary to give the general

plan of the superficial origin of the preganglionic

fibres from the central nervous system. Detailed

description will be given later.

The broad facts with regard to the origin of the

tectal, bulbar, and sacral nerve fibres have long been

known. The ciliary nerves arise from the region of

the anterior corpus quadrigeminum and form part

of the 3rd cranial nerve. The bulbar nerves arise

from the lower part of the spinal bulb and issue in

the portio intermedia of the 7th cranial nerve, and

in the 9th, loth and bulbar part of the nth nerves.

The 5th nerve has been said to send inhibitory fibres

to the sphincter of the pupil and vaso-dilator fibres to

the mouth, but if any effect is produced by this nerve

it is probably by antidromic action. The sacral

autonomic nerves arise from the middle and lower

part of the sacral spinal cord and issue in sacral nerves

(generally three) numbered differencly in different

animals.

The determination of the origin of the preganglionic

fibres of the sympathetic is more recent. At the time

of Gaskell's researches the sympathetic was supposed

to arise from all the spinal nerves. The anatomical

observations of Beck on man (1846), if they had

received attention at a later time, should have raised

doubt as to the origin of sympathetic fibres from the

lower and upper regions of the spinal cord. He des-

cribed the thoracic and lumbar nerves as having both

15

1

6

AUTONOMIC NERVOUS SYSTEM

white and grey rami communicantes, the white rami

becoming smaller in passing down the lumbar region,

and the cervical and sacral nerves as having grey rami

communicantes only. But at this time little was

known with certainty either of the origin of the

"tubular" fibres which formed the chief constituent

of the white rami, nor of the " gelatinous " fibres which

formed the chief constituent of the grey rami. The

question discussed was the degree of independence of

the gangha. Beck supported the independence

theory, and the bearing of his results on the connexion

of different parts of the spinal cord with sympathetic

ganglia passed unnoticed. Reissner (1862) described

bundles of small nerve fibres as being present in the

anterior nerve roots of the dorsal region of the cord,

and scattered small fibres only as being present in

those of the cervical and lumbar regions, but this was

not correlated with the existence of small fibres in

the sympathetic.

So far as the thoracic and upper lumbar nerves were

concerned the experimental evidence was ample to

show that they influenced either viscera or blood

vessels by way of the sympathetic. The evidence as

regards the cervical, lower lumbar and sacral nerves

was conflicting. If we take the number of observers

who described a result as a criterion of the probability

of its truth, the balance of evidence was strongly in

favour of the cervical nerves sending fibres to the

cervical sympathetic. Budge, Frangois-Franck,

Salkovski, Navrocki and Przybylski had all described

pupillo-dilator fibres as arising from one or more of the

lower cervical nerves. CI. Bernard ( 1 862) almost alone

found no pupil effect from the cervical nerves. Dastre

and Morat described the lower cervical nerves as

s

ORIGIN AND PERIPHERAL DISTRIBUTION 17

sending vaso-constrictor fibres to the cervical sympa-thetic. Bever and Bezold, Boehm and Nussbaum,who investigated the origin of sympathetic cardiac

accelerator fibres, considered them to arise in part

from the lower cervical nerves. On the other handthe evidence was against the lower lumbar or sacral

nerves sending either vaso-motor or secretory fibres

to the limbs by way of the sympathetic, and the

question at issue was mainly the presence or absence

of vaso-motor and secretory fibres taking a direct

course in the spinal nerves. The lower lumbar and

sacral nerves were supposed to send fibres to the

viscera by way of the sympathetic chain, but there

was no satisfactory evidence for this.

Gaskell (1886) approached the subject from a

histological and morphological standpoint. He found

in the dog that bundles of small nerve fibres i -8 to

3'6/x in diameter occurred in the anterior roots of

those nerves which had white rami, but not in the

anterior roots of the nerves which had grey rami,

except in the case of the sacral nerves from which the

nervus erigens arose, and this had no connexion with

the sympathetic chain. The white rami—as had been

shown by Bidder and Volkmann and others—con-

sisted mainly of similar small medullated fibres. Ontracing the grey rami towards the spinal cord he found

no non-medullated fibres in the roots, inside the dura

mater; the few non-medullated fibres found more

peripherally he considered ran to the dura mater and

thence to the blood vessels of the cord.

A somewhat similar result and conclusion had beenarrived at by Beck in man. He described a few "gela-

tinous" fibres in the roots, not traceable to the cord, butseeming to run to the blood vessels. (Todd and Bowman'sPhysiological Anat. and Physiol, of Man, 2, p. 132, 1859,

1

8

AUTONOMIC NERVOUS SYSTEM

London.) The statement of Ranson that there are numer-ous non-medullated fibres in the posterior roots I shall

discuss later.

From these and other results Gaskell came to the

important conclusion that only those nerves which

had white rami sent fibres to the sympathetic system.

These formed his "thoracic outflow." In the dog he

found white rami from the 2nd thoracic to the 2nd

lumbar (the 4th lumbar of most writers). The hmits

of the white rami will be dealt with under the section

of the sympathetic system ; here it is sufficient to take

them as present in the dog from the ist thoracic to

about the 4th lumbar nerve. The facts described by

Gaskell showed conclusively that the very great

majority of the nerve fibres passing to the sympathetic

arose from a limited region of the spinal cord. They

were less conclusive as to the total absence of such

fibres from the upper and lower regions of the cord.

As evidence of total absence Gaskell relied on two ob-

servations, first that the anterior roots of the cervical

and lower lumbar nerves had no fibres less than 3-6;u in

diameter. This statement is not quite accurate ; there

are a few fibres 2 to 3'6/x, i.e. of the size of sympathetic

preganglionic fibres, in the anterior roots of the nerves

having no white rami. The other point relied on was

that the grey rami contained very few small medul-

lated fibres. In fact, however, the grey rami of the

cat and of some other animals contain a considerable

number of small medullated fibres (cp. p. 24). Whetherthere are few or many it is impossible to follow themwhen undegenerated through the nerve trunk and into

the roots in such way as to be certain that none arise

from the spinal cord.

The contradiction between the conclusion drawn

ORIGIN AND PERIPHERAL DISTRIBUTION 19

from histological observations and those drawn fromexperiments for the most part disappeared as further

experiments were made on the effect of stimulating

the spinal nerves at their origin. (These experiments

will be considered later under the headings of the

"sympathetic" and of "antidromic action.")

There was still the possibility that the experimental

method was not sufficiently delicate to detect the

trifling effect which would be produced if a few fibres

ran from the spinal cord to the sympathetic by wayof the grey rami. Evidence on this point I obtained

by the degeneration method (1896). On cutting the

lumbar and sacral nerves inside the vertebral canal,

either no small medullated fibres were found in the

grey rami, or a few were found which there was good

reason to believe arose from white rami or were

afferent fibres. The function of the small nerve fibres

in the anterior roots of the nerves which have no

white rami remains to be determined.

The origin of the preganglionic fibres from the spinal

cord may then be diagrammatically represented as

in Fig. I.

Fig. I. Tectal Bulbar Thoracico-lumbar Sacral

The general plan of the peripheral connexion of the

several parts of the autonomic nervous system is, as

has been said above, that the thoracico-lumbar or

sjnnpathetic system supplies nerve fibres to all

regions of the body, whilst the tectal, the bulbar and

the sacral systems supply nerve fibres to special

regions only.


The tectal or mid-brain system supplies the sphinc-

ter of the iris and the ciUary muscle.

The bulbar autonomic supplies almost exclusively

the parts which have developed from or in connexion

with the primitive alimentary canal. It innervates

the mucous membrane of the nose, mouth and

pharynx with the lachrymal and salivary glands, the

oesophagus, stomach, liver, pancreas, the small

intestine, and in some animals at any rate the anterior

part of the large intestine. It innervates also the

lungs (which develop as a diverticulum from the

oesophagus) and the heart.

The sacral autonomic innervates the lower part of

the large intestine, possibly also the upper part, the

bladder, and the external generative organs (penis

and vagina).

Whilst anatomically the sympathetic runs to all

regions of the body, and the parasympathetic to someof these regions, it does not follow that either system

of nerves has an appreciable effect on all the structures

in the regions to which it runs. Thus, whilst both

sympathetic and parasympathetic nerves run to the

eye, it is an open question whether any part of the

eye is innervated by both nerves ; the bulbar system in

some lower vertebrates is said not to innervate the

ventricle of the heart, and there is no satisfactory

evidence that the intestinal glands are innervated byeither bulbar or sympathetic nerves. The details of

the connexions I shall consider under the head of the

separate nerve systems.

ORIGIN AND PERIPHERAL DISTRIBUTION

REFERENCES.

Beck. Phil. Trans. Roy. Soc. London, p. 213, 1846.

Reissner. A. f. Anat. u. Physiol., p. 125, 1862.

Claude Bernard. J. de la Physiol. 5, p. 383, 1862.

Gaskell. J. Physiol. 7, p. i, 1886.

Langley. Ibid. 20, p. 223, 1896.

References on the origin of sympathetic fibres from the

spinal cord which are quoted in the text are given in most of

the larger, not too recent, text-books. References up to 1903are given in my article in the Ergebn. d. Physiol. 2, Abth. 2,

p. 820; since I shall refer to the work from 1889 onwards underthe "sympathetic system," the references need not be re-

peated here. The opinions held in 1879-80 may be gathered

from Funke and Griinhagen's Lehrbuch d. Physiol. 6te Aufg.

2, p. 713, 1879, 3nd from Hermann's Hdb. d. Physiol. 2, Th. i,

p. 275, 1879; 4 Th. I, p. 343, 1880.

3. The Nerve Fibres of the Autonomic

System.

In the early observations which were made on the

nerves it was noticed that certain parts of the inter-

costal nerve were reddish or greyish, whilst other

parts and all the spinal nerves were white. Out of

these observations arose the terms "grey and white

rami communicantes " of the spinal nerves. Whenthe microscope came into use, the white nerves were

found to contain numerous fibres which were called

tubular—the meduUated or myelinated of to-day.

They were considered to be the only form of nerve

fibre, and it was noticed that in the sympathetic

nerves the great majority of tubular fibres were small.

Remak (1837-8) found a different form of nerve fibre

which he called "organic" on the lines of Bichat's

theory. It is the non-medullated or non-myelinate d

fibre of to-day. He considered that it was the sole

form of nerve fibre given off by the sympathetic and

spinal ganglia, and that the cerebro-spinal fibres were

all tubular. Bidder and Volkmann (1842), on the

other hand, after a very thorough examination of a

large number of nerves in different vertebrates, cameto the conclusion that the sympathetic and spinal

ganglia gave off most and probably all the small

tubular nerve fibres of the body, that the brain and

spinal cord gave off all the medium and large fibres,

and that Remak 's fibres were not nerve fibres. Thegeneral recognition that the "organic" fibres were

nerve fibres was slow. Those who accepted it

NERVE FIBRES OF AUTONOMIC SYSTEM 23

necessarily rejected Bidder and Volkmann's theory.

Kolliker (1844) confirmed Bidder and Volkmann's

statement that some small tubular fibres arose from

the ganglia, but contested the theory that they were

confined to the ganglionic system. This theory

indeed received little support from anyone. The

nerve fibres arising from the posterior root ganglia

were eventually recognised as being almost wholly

medullated, but it was a not uncommon belief that

some of them were non-medullated.

Discussion of distinguishing differences between

nerves running to ganglia, those arising from them,

and those running direct to the tissue, died out. It

is an indication of this that the question was not even

mentioned by Max Schulze in his account of nerve

fibres in Strieker's Histology in 1870.

After a long interval a return to generalisation was

made by Gaskell (1886). He put forward the theory

that all the efferent "visceral" nerves passing from

the central nervous system were small medullated

fibres, that thej^ lost their medullary sheath in the

ganglia, and that the ganglia gave off non-medullated

fibres. The simplicity of this theory and its obvious

truth in. the case of many nerve fibres in the mammalled to its obtaining a considerable degree of acceptance.

The theory, it will be seen, consists of three parts.

The first, viz., that all efferent "visceral" fibres as

they leave the brain or cord, i.e. all pre-ganglionic

fibres are small medullated fibres has been confirmed

or not contested by subsequent observers. The

second and third parts of the theory, viz., that the

pre-ganglionic fibres lose their medulla in the gangha,

and that the gangha give off non-medullated fibres,

and these only, was important, since, if true, a means


was afforded of allowing preganglionic to be dis-

tinguished from postganglionic fibres and of determin-

ing whether a ganglion had sympathetic nerve cells

in it. Unfortunately the matter is less simple.

In the rabbit, as I pointed out in 1892, the great

majority of the preganglionic fibres which run to the

sacral and coccygeal ganglia lose their medulla a

considerable distance before they reach the ganglia

in which they end. In all mammals a loss of medulla

appears to occur in some of the corresponding fibres,

and in the majority of the preganglionic fibres of the

splanchnic and vagus nerves. As to the last part of

the theory, viz., that the ganglia do not give off

medullated fibres, Gaskell a little later (1889) noticed,

as had been noticed by Bidder and Volkmann, that

the short ciliary nerves were medullated, so that the

theory did not hold for the tectal autonomic. Nor

does it hold for the sympathetic nerves of the dog and

cat. In the cat I found (1896) the 7th lumbar grey

ramus to have more than 300 medullated fibres, and

the anterior branches of the superior cervical ganglion

to have several hundreds, and showed by the degene-

ration method that they were nearly all postganglionic.

Billingsley and Ranson (191 8) counted in different

cats about 1 500 to 4000 small medullated fibres in the

grey rami given off by this ganglion. In the grey

rami of the dog the medullated fibres are fewer; in

several thoracic rami I found 10 to 25 only, but there

were many in the lumbar rami. In the rabbit the

medullated fibres of the grey rami are few, and

possibly in this animal none of them are postganglionic

;

but as the preganglionic are apt to become non-

medullated, the two classes of fibres cannot be

distinguished with certainty. The observations of


Bidder and Volkmann showed that in the frog, and

probably in fish, reptiles and birds the ganglia gave

off numerous small meduUated fibres. In fact, in

vertebrates other than mammals, the rami com-

municantes of the spinal nerves consist wholly, or

almost wholly, of meduUated fibres; grey rami com-

municantes are not present. On the other hand,

most of the postganglionic fibres passing to the

abdominal viscera in lower vertebrates are non-

medullated.

The meduUated postganglionic fibres of the mammalas a rule, lose their medulla at some point on their

course to the periphery, and I am not certain that

any remain meduUated up to their final branches.

Sherrington (J. Physiol. 17, p. 248, 1894) did not find anymeduUated degenerated fibres in the muscular and cutaneousnerves of the cat after excision of the lumbar and uppersacral ganglia in the cat. Some of the branches of the

sciatic in the frog contain a bundle of non-meduUated fibres,

with a few small meduUated in them, though the post-

ganglionic fibres entering the sciatic are meduUated.

No difference in function has been found between

meduUated and non-medullated postganglionic fibres.

The hypothesis I would suggest as to the proximate

cause of the existence of the two kinds of nerve fibres

is that cells with non-medullated fibres were the first

in phylogeny to migrate from the central nervous

system, a later migration occurring when a further

specialisation of the central nervous cells had oc-

curred, and that the cells of this migration gave rise

to meduUated fibres. On this hypothesis, the two

forms of embryonic cells have persisted to a varying

degree in different vertebrates, each form giving rise

to its own kind of axon.

Gaskell gives i-S to 3-6^ as the size of the small


fibres in osmic acid preparacions. Some preganglionic

fibres are a trifle larger than this—up to 4/x in diameter.

The sympathetic postganglionic fibres are rarely

larger than 2-S/ui; the amount of myelin in them varies;

in the mammal, at any rate, it may form a very thin

layer requiring deep staining to make it obvious. In

transverse sections of grey rami stained with osmic

acid, fibres with very thin sheath may easily be over-

looked.

There is much that is still unknown with regard to the

sheath of the non-medullated fibres. In the sacral region

the preganglionic nerve fibres which have lost their myelinare not obviously different from postganglionic fibres whichhave none. Yet the sheath of a postganglionic fibre haslittle resemblance to the sheath (neurilemma) of a medul-lated fibre. Tuckett describes the postganglionic fibres

near their origin as having a relatively thick sheath whichreadily breaks up into fibrillse, in which the nuclei are

imbedded. But at the periphery the non-medullated fibres

forming the terminal plexus appear to be without sheath.

Where the sheath is lost, if it is lost, is unknown. Thesheath of the non-medullated fibres of the thoracic vagusseems to me to be different from that of the sympatheticfibres.

After section of postganglionic nerves the nerves

cease to be irritable in a few days, but when teased

out in osmic acid preparations the non-medullated

fibres may maintain their usual appearance for two

or more weeks. Tuckett (1896) found that the central

axon core ceased to stain normally with methylene

blue about the time the irritability disappeared.

Recently non-medullated fibres have been described

by Ranson (191 2) and by Boeke and Dusser de

Barenne (1919) as degenerating very much later than

medullated fibres. They treated the nerves byBielschowsky's silver method, and I consider that

what they took for undergenerated non-medullated

fibres was only the sheath of the fibres.


In the preceding account I have dealt with certain

points common to all autonomic nerve fibres. Somefurther account of the histological characters of the

grey and white rami will be given under the sympa-

thetic system.


Remak. Froriep's Notizen, 1837. Observationes anat. et

microc. de systematis nervosi structura (Berol), 1838.

Bidder and Volkmann. Die Selbstandigkeit d. Symp.Nervensystems (Leipzig), 1842.

Kolliker. Die Selbstandigkeit and Abhangigkeit d. Symp.Nervensystems (Zurich), 1844.

Gaskell. J. Physiol. 7, p. i, 1886. Ibid. 10, p. 153, 1889.

Langley. Phil. Trans. London, 183, B. p. 85, 1892.

Tuckett. J. Physiol. 19, p. 267, 1896.

Langley. Ibid. 20, p. 55, 1896.

Ranson. J. Comp. Neur. 22, p. 487, 1912.

Billingsley and Ranson. Ibid. 29, p. 367, 1918.

Boeke and Dusser de Barenne. Proc. k. Akad. Am-sterdam, 21, 1919.

Numerous small nerve fibres are present in the posterior

roots of all spinal nerves; whether these are especially con-

cerned with antidromic action is unknown.

4. The Specific Action of Drugs on the

Sympathetic and Parasympathetic

Systems. ^

One reason for dividing the autonomic nerves into

sjrmpathetic and parasympathetic is, as I have said,

that the sympathetic nerves supply all parts of the

body, whilst the remaining nerves supply special

parts only. And this division is convenient in

describing pharmacological results. Adrenaline and

certain related substances produce effects which are

in nearly all cases like those produced by stimulating

sjonpathetic nerves. Pilocarpine, muscarine and

arecoline, on the other hand, whilst having some

differences in action, produce effects which are in most

cases like those produced by stimulating parasympa-

thetic nerves. Most of the effects produced by

physostigmine and by choline are such as would be

produced by increasing the excitability or by stimulat-

ing parasympathetic nerves. Atropine abolishes the

effects of the pilocarpine group of drugs, dulls, and in

some cases abolishes the effect of stimulating the

parasympathetic nerves. The facts show that there

is a close relation between the action of the drugs and

innervation by sympathetic and parasympathetic

nerves respectively, and they suggest that there is a

fundamental difference between the two systems.

The facts are also of importance with regard to the

^ I discuss this subject before that of the tissues innervated by thesympathetic chiefly because the action of certain drugs is commonlytaken as showing whether a tissue is innervated by bulbo-sacral orsympathetic nerves, and to give the evidence here will save repetition.

28

SPECIFIC ACTION OF DRUGS 29

action of hormones on the tissues. It is generally held

that the adrenaline formed in the body sustains

sympathetic action ; and it has been held that choHne,

which is also formed in the body, sustains para-

sympathetic action.

Taking adrenaline and pilocarpine as respectively

representing the two classes of drugs, we may con-

sider the cases in which there is, or appears to be, a

lack of correspondence between the action of the drugs

and that of nerve stimulation.

Barger and Dale (J. Physiol. 41, p. 19, 1910) use the

term sympathcmimetic to describe the action cf amines.

This term, and paxasympathomimetic may conveniently beused to imply that the effects produced by the drugs

resemble broadly, but not exactly, the effects producedby nerve stimulation.

Adrenaline and sympathetic effects.

The only structures markedly influenced by sympa-

thetic nerve stimulation which are not influenced by

adrenaline after nerve section are the sweat glands of

the cat and of some other mammals. Dieden (19 16)

observed that injection of adrenaline into the sub-

cutaneous tissue of the cat's foot in certain conditions

caused secretion, but this, I find (1921), is due to the

fluid in which it is dissolved, and not to the adrenaline

itself. A doubtful difference in action is presented

by the statements regarding the tone of striated

muscle. On the basis of the effects of nerve section,

the sympathetic is held by some observers (cp. p. 71)

to cause tonic contraction. This has not been found

to be produced by adrenaline, but it has also not been

found to be caused by sympathetic nerve stimulation.

When the relative effect of sympathetic stimulation

and of adrenaline on different tissues is compared,


there are some striking instances of lack of corres-

pondence. Thus weak stimulation of the sympathetic

in the cat causes contraction of the cutaneous arteries,

contraction of the tunica dartos of the scrotum and of

the erector muscles of the hairs. A very small amount

of adrenaline causes contraction of the cutaneous

arteries, a larger amount is required to cause con-

traction of the tunica dartos, and a very much larger

amount is required to cause contraction of the erector

muscles of the hairs.

There is a similar difference in the sympathetic and the

adrenaline effects on the arteries and on the unstriated

muscles causing movement of the feathers of the fowl(Langley, J. Physiol. 30, p. 240, 1903). The relatively

slight effect on skin muscles is not, however, universal, for

adrenaline readily causes erection of the quills of the hedge-hog (Lewandowsky, igoo), of the hairs of the tail of the

marmot (Kahn, A. (Anat. u.) Physiol., p. 239, 1903), of thehairs of the mongoose, and goose skin in man (Elliott, 1905).Elliott connected the degree of action of adrenaline with thenormal frequency of sympathetic stimulation.

There is also some lack of correspondence in readiness

of production of contraction and inhibition bysympathetic nerves and by adrenaline when the

sympathetic has both contractor and inhibitory

fibres. The most striking instance is that of the

bucco-facial region of dog, in which adrenaline has not

been found to cause any flushing but only contraction.

In the cat's bladder on the other hand, adrenaline

causes inhibition more readily than does the sympa-

thetic.

In the bladder of the cat, Langley (1901) and Elliott

(1905) found no preliminary rise of internal pressure withadrenaline such as is caused by sympathetic stimulation.Edmunds and Roth (J. Pharm. exp. Ther. 15, p. 189, 1920)usually obtained a preliminary rise, but on the excisedtrigonal region—the part which contracts on sympatheticstimulation—they frequently failed to get contraction.


Adler (A. exp. Path. Pharm. 83, p. 248, 1918) usually-

obtained inhibition of the excised frog's bladder withadrenaline; the ordinary effect of sympathetic stimulationis contraction. The different effects may be due to differ-

ences of dose, but this has not been shown.

Further, some effects have been obtained with

adrenahne which either have not been obtained by-

sympathetic stimulation or which have not been

obtained with certainty. More decided effects on

cerebral vessels have been obtained by adrenaline

than by sympathetic stimulation. Except in the

nictitating membrane of the frog, no certain effect of

the sympathetic on capillaries has been described, but

there is some evidence that adrenaline may cause

either contraction or dilatation of capillaries. Capil-

lary dilatation may be the cause of the fall of blood

pressure which a small amount of adrenaline produces

in carnivora; the fall has, however, been usually

attributed to an action on arteries, and this also has

not been obtained in normal conditions by sympa-

thetic stimulation. Adrenaline has also been said to

cause a slight movement of pigment granules in the

deep-lying pigment cells of the frog, an effect not

produced, so far as is known, by sympathetic nerve

stimulation (cp. p. 58).

Fall of blood pressure. Dilatation of blood vessels. Mooreand Purinton found that suprarenal extract in minimalamount caused a fall of blood pressure in dogs (A. ges.

Physiol. 81, p. 483, 1900), and Hoskins and McClure showedthat the effect was due to adrenaline (A. Int. Med. 10,

p. 353, 1912), see also Cannon and Ljnnan (Amer. J. Physiol.

31, p. 376, 1913). The amount required to cause a fall of

blood pressure varies in different tissues (Hartman, ibid. 38,

p. 438, 1915, and other papers in Vols. 43 and 44). Adrena-

line has not been found to cause fall of pressure in the rabbit,

but Ogawa (A. exp. Path. u. Pharm. 67, p. 89, 1912) states

that perfusion of the intestine, kidney and skin vessels of the

rabbit with very dilute adrenaline causes dilatation when the


perfusion is kept up for some time (cp. also Bauer and Froh-

lich A. exp. Path. u. Pharm. 84, p. 33, igo8). Desbouis

and Langlois state that in minimal dose its effect on the

vessels of the lung is dilatation (C.R. Soc. Biol. 72, p. 674,

1912), and Rothlin (HabUitationsschrift Ziirich, 1920) finds

a similar effect on the isolated renal artery.

Cerebral vessels. Biedl and Reiner (A. ges. Physiol. 79,

p. 193, 1900) and Wieschovski (A. exp. Path. u. Pharm. 52,

p. 389, 1905) came to the conclusion that adrenaline caused

strong contraction of the cerebral vessels. Dixon andHalliburton, in perfusion experiments, found slight dilata-

tion only which they thought was confined to the larger

arteries (Quar. J. exp. Physiol. 3, p. 315, 1910), Wiggersobtained contraction (Amer. J. Physiol. 14, p. 452, 1905;

J. Physiol. 48, p. 109, 1914). Cow in ring preparations of the

cerebral arteries found slight dilatation (J. Physiol. 42,

p. 125, 1911).

Veins. According to Gunn and Chavasse, adrenaline

causes rhythmic contraction in ring preparations of the

superior vena cava and contraction of the jugular veins of

the sheep (Proc. Roy. Soc. B. 86, p. 192, 1913). Doneganstates that in the cat adrenaline in perfusion experiments hasno effect on the large veins near the heart (J. Physiol. 55,

p. 226, 1921).

Capillaries. On local application of adrenaline to the

intestine I found intense pallor, and suggested that it wasdue to contraction of the capillaries (J. Physiol. 27, p. 248,

1901). Protopopov (Abstract in J. de Physiol. Path, gen.,

p. 1122, 1904), on injecting adrenaline, observed pallor of

the brain, which he thought was probably due to a direct

action on the capillaries since the visible arteries did notcontract. Cotton, Slade and Lewis obtained pallor in theforearm on local injection of adrenaline in man after

stopping the blood flow by compression of the upper arm(Heart 6, p. 246, 1917). Dale and Richards describeadrenaline as causing marked dilatation of capillaries in

carnivora (J. Physiol. 52, p. no, 1919). Krogh obtaineddilatation of capillaries in the tongue of the frog on local

apphcation of adrenaline (J. Physiol. 53, p. 408, 1920).Elliott, on local apphcation of adrenaline to the capillaries

of the nictitating membrane of the frog, found no effect

(J. Physiol. 32, p. 414, 1905) ; these are the only capillaries

which have been found to contract on sympathetic stimula-tion (cp. p. 64). The vaso-dilation caused by histamine,which Dale and Richards attribute to capillary action, is

said by Schenk (A. exp. Path. u. Pharm. 89, p. 332, 1921)to be prevented by adrenaline.


Adrenaline has been described as having on striated

muscle several effects which have not been described

as occurring on sympathetic nerve stimulation. All

recent evidence tends to show that adrenahne in the

amount at all likely to be present in the blood has noeffect on the height or duration of the single muscular

contraction, but it tends to show that in some amountbeyond this it has an action similar to that of fatigue.

How far this action is due to the substances which

are mixed with adrenaline in the preparations used

has not been clearly settled. There is, however,

much evidence that when a muscle has been fatigued,

a partial restoration is brought about by dilute

solutions of adrenaline (Dessy and Grandis and others),

and there is some evidence that adrenaline partially

antagonises the action of curari. The effect on

fatigue has been attributed by most observers to an

action of adrenaline throughout the muscle fibres,

and not to an action confined to the myo-neural

junction, or to the neural region of the muscle fibres.

Other instances of apparent lack of correspondence

between the effects of adrenaline and of sympathetic

stimulation are mentioned below under "reversed

action" (cp. p. 37).

Action on muscle. Oliver and Schafer (1895) described anincrease in height and duration of the single contraction of

muscle (frog and dog) as being caused by suprarenal extracts.

Boruttau (1899) found both by indirect stimulation and bydirect stimulation of the curarised frog's gastrocnemius

muscle that adrenaline caused effects similar to those caused

by fatigue. Panella (A. Ital. de Biol. 48, p. 430, 1907)obtained no effect on the height of contraction of excised

frog's muscles, or on normal mammalian muscle. Kunofound (1915) that -006 p.c. adrenaline had no effect on the

curve of single muscle contraction of the frog's sartorius

muscle. An absence of effect was also noted by Takayasu(Quar. J. exp. Physiol. 9, p. 347, 1916) with '0001 p.c.


adrenaline. Schafer (Endocrine Organs, p. 65, 1916)attributed his earlier results to substances other thanadrenaline in the extracts. Takayasu described a lessened

height and duration of contraction on immersing the frog's

sartorius in adrenaline -0002 p.c. and upwards. Okushiraa(Acta Schola med. Kioto 3, p. 251, 1919), on indirect stimu-

lation of the gastrocnemius of the frog, obtained increased

height of contraction with adrenaline -oooi p.c, and nerveparalysis with -oi p.c. (The Parke, Davis preparation usedcontains chloretone, and the effect of this was not tested).

On direct stimulation of the muscle he found only a slight

decrease with -oi p.c. adrenaline.

Dessy and Grandis (A. Ital. de Biol. 41, p. 225, 1904)found in a frog-toad (Leptodactylus) that injection of

adrenaline partially restored the height of muscular con-

traction which had been reduced by a series of single induc-

tion shocks. A similar but less effect was found on adding

adrenaline to excised muscles. Panella (Ibid. 48, p. 430,

1907), in frogs, toads, rabbits and guinea pigs, confirmed aneffect of adrenaline on fatigued muscles. Cannon and Nice

(Amer. J. Physiol. 32, p. 44, 1913) showed that the effects

previously produced were largely due to circulatory changes,

but gave additional evidences of a direct action of adrenaline

on muscular contraction. The most complete experimentson the mammal are those of Gruber (Ibid. 33, p. 358; 34,

p. 89, 1914). Gruber and Fellows (Ibid. 46, p. 472, 1918)

and on amphibia by Guglielmetti (Quar. J. exp. Physiol. 12,

p. 139, 1919). Other references are given in these papers.

A recovery from fatigue produced by direct stimulation of

the muscle has been found on giving adrenaline by Gruberin the curarised cat's muscle and by Guglielmetti in the

frog's gastrocnemius. Okushima did not find that adrena-

line had an effect on the fatigue of the frog's gastrocnemius.

The experiments so far made do not definitely decide

whether the action of adrenaline on fatigue is due to anaction throughout the muscle or not. In many of theexperiments the nerve fibres may have been stimulated.

Boruttau and Gruber found an effect after curari, but it is

possible that the amount of curari did not reduce theexcitability of the neural region of the muscle to that of thenon-neural region.

Panella (A. Ital. de Biol. 47, p. 17, 1907) and Gruber (op.

cit. vol. 34) describe adrenaline as partially antagonising theparalysing action of curari.

Some other observations on the effect of adrenaline onstriated muscle I give on p. 76.


Lastly, one action of adrenaline has been attributed

to stimulation of the end apparatus of parasympatheticnerves. Luckhardt and Carlson (1921) find that

whilst adrenaline causes contraction of the pul-

monary arteries of the frog and turtle, sympathetic

stimulation does not. They find that in these

animals the vaso-constrictor fibres for the lung run in

the vagus and are paralysed by atropine.

The account given above may be summarised as

follows:—^There is only one clear case in which

sympathetic stimulation has a marked effect andadrenaline has none (the sweat glands), but the degree

of effect of the two forms of stimulation do not run

parallel in the several tissues, the balance of con-

tractor and inhibitory effect is not the same, and one

or other effect caused by sympathetic stimulation

may apparentty be absent on injecting adrenaline.

Conversely adrenaline has an effect in some cases in

which stimulation of the sympathetic has not been

shown to have any; in general these are cases in

which very dilute adrenaline is used. In one case

(lung vessels of frog and turtle) adrenaline is said to

have the same effect as parasympathetic nerve

stimulation.

Pilocarpine and parasympathetic effects.

Nearly all the effects which are caused by stimula-

tion of parasympathetic nerves are also caused by

pilocarpine. The most marked exception is that the

sacral autonomic nerves cause inhibition of the

retractor penis, and pilocarpine causes contraction.

The degree of effect produced by the two forms of

stimulation is not always the same; it differs strik-

ingly in the small intestine. The action of pilocarpine


is not confined to the tissues innervated by para-

sympathetic nerves, it causes secretion from sweat

glands, contraction of the uterus, and contraction

or dilatation of arterioles. It causes also contraction of

the spleen, an effect which has not been found to be

produced by stimulation of the vagus. Similarly, the

paralysing action of atropine is not confined to para-

sympathetic nerves, it paralyses the sympathetic

secretory fibres of the submaxillary gland, and of the

sweat glands of the cat, the secretory nerves of the

skin glands of the frog, and, according to Mislavski

and Borman, the sympathetic secretory fibres of the

prostate gland.

Uterus. Cushny (J. Physiol. 41, p. 238, 1910) found that

in the cat injection of pilocarpine caused an effect similar to

that caused by sympathetic stimulation, i.e. as a rule

inhibition in the non-pregnant animal and contraction in the

pregnant animal. Dale and Laidlaw (Ibid. 45, p. i, 1912)found the inhibition to be inconstant and not to occur in the

excised uterus; they attributed the inhibition in the bodypartly to a liberation of adrenaline and partly to stimulation

of sympathetic ganglia by pilocarpine. The uterus of the

guinea-pig, whether pregnant or non-pregnant, they foundto be contracted only by pilocarpine. Gunn and Gunn,however (J. Pharm. exp. Ther. 5, p. 527, 1914) sometimesobtained inhibition in the excised uterus of the guinea-pig,

and in two experiments inhibition in that of the rat, thoughmore often there was no decided effect. Dale and Laidlawdidnotfindthecontractoraction of pilocarpine to be abolished

by ergotoxine. According to Gohara (Acta Sch. Med. Kioto.

3, p. 363, 1920) atropine abolishes the contractor effect of

adrenaline on the uterus of the rabbit.

For contraction of the retractor penis by pilocarpine cp.

Bottazzi (A. Ital. de Biol. 65, p. 265, 1916). Edmunds(J. Pharm. exp. Ther. 15, p. 201, 1920) found that pilocar-

pine still caused contraction after ergotoxine had abolished

the motor effect of adrenaline. Contraction of the spleen is

mentioned by Harvey (J. Physiol. 35, p. 116, 1906) and byDixon (Manual of Pharm., p. 88, 1919).

Blood Vessels. Pilocarpine causes flushing of the skin of

man. And experiments on animals indicate that on injec-

tion it has a peripheral vaso-dilator action. (Langley,


J. Anat. and Physiol. 10, p. 66, 1875; Reid Hunt, Amer. J.Physiol. 44, p. 254, 1918.) According to Brodie and Dixon

(J. Physiol. 30, p. 485, 1904), pilocarpine causes contraction

of vessels in the intestine and limbs in perfusion experiments.It has been said to cause slight dilatation of lung vessels

(Brodie and Dixon, op. cit. supra., p. 488; Baehr and Pick,

A. exp. Path. u. Pharm. 74, p. 55, 1913), and also contrac-

tion (Beresin, A. ges. Physiol. 158, p. 219, 1914). Dixonand Halliburton (Quar. J. exp. Physiol. 3, p. 315, 1910)found that it caused dilatation of the vessels of the brain.

Amsler and Pick did not find that it had any effect on the

perfused vessels of the frog (A. ges. Physiol. 85, p. 61, 1919).Dixon (Manual Pharmacol., 4th edit., p. 86, 1919) states

that it stimulates the end apparatus of the accelerator nerve.

There are other instances of a difference of effect in pilo-

carpine and parasympathetic nerve stimulation.

Paralysis of sympathetic fibres by atropine. Langley,

Submaxillary gland of the cat (J. Physiol, i, p. 97, 1878;Strieker and Spina, Skin of glands of the frog (Med. Jahrb.,

P- 355. 1880) ; Mislavski and Borman, Prostate gland

(Centrlb. Physiol. 12, p. 181, 1898).

Reversal of effect of sympathetic and parasympathetic

drugs.

In the preceding account I have mentioned some

cases in which a very small amount of adrenahne

produces effects different from those caused by larger

amounts and effects which have not been produced,

or only doubtfully, by nerve stimulation. These are

often spoken of as a reversal of adrenaline action so as

not to prejudge the question whether they are

normally produced in the body by nervous action or

not. In perfusion experiments on the frog's heart

and blood vessels, reversed action has been obtained

by varying the amount of Ca salts in the perfusion

fluid, by making the Ringer's fluid slightly acid, by

previous treatment with another drug and by pro-

longed action of adrenaline. Whilst in some cases

the reversal has been taken as showing the presence


of a small number of nerve fibres antagonistic in

action to that of the majority, in others the reversal

has been taken to show that the action is determined

by the condition of the tissue and is not an expression

of normal nerve action, and in others again it has

been taken as showing an action of a "sympathetic

drug" on parasympathetic end apparatus, or of a

"parasympathetic drug" on sympathetic end appa-

ratus. One difficulty in interpreting the results lies

in determining how far they are due simply to the

displacement of one adsorbed substance by another.

As they concern us here, they give examples in which

a relation of drug action to nerve systems has not been

shown.

The reversed action of adrenaline after ergotoxine,

since it is commonly held to be due to stimulation of

sympathetic inhibitory nerves, will be considered under

the head of the sympathetic system.

Pearce (Zts. f. Biol. 62, p. 243, 1913) found in perfusion

experiments on the vessels of the frog 4, 6, and 29 days after

the sciatic had been cut, and also after perfusion with pureNaCl, that a small amount of adrenaline caused vaso-

dilation provided curari had been given, and occasionally

without it. Diastolic cessation of the heart beat of the frog

by adrenaUne was obtained by Burridge after perfusion with

3 p.c. sodium phosphate (Quar. J. exp. Physiol. 5, p. 368,,

1912) and by Kolm and Pick after decrease of calcium in

the perfusion fluid (A. ges. Physiol. 184, p. 79, 1920). Kolmand Pick also found that adrenaline stopped the heart in

diastole when given after a certain amount of acetylcholine

;

they attributed the effect to stimulation of the end apparatusof vagus fibres since it was prevented by atropine. Snyderand Campbell (Amer. J. Physiol. 47, p. 199, 1920) observedchange of the action of very dilute adrenaline from vaso-

constriction to vaso-dilation when the fluid was made slightly

acid.

Kato and Watanabe after repeated injection of adrenaline

in the dog, found that adrenaline placed on the conjunctiva,

caused contraction of the pupil (Tokio J. exp. Med. i, p. 73^


1920). This, however, may possibly have been due to anincreased responsiveness of the blood vessels; it was foundby myself and by Elliott that dilatation of the pupil in thedog requires a relatively large dose of adrenaline. Afterperfusing the frog's heart with Ringer's fluid containingadrenaline and excess of Ca, Pick found that acetyl-choline

—

a parasympathetic drug—caused contracture instead of

inhibition ; the effect was not prevented by atropine, but wasprevented by ergotoxine, and Pick considered it to be dueto an action of acetyl-choline on the end apparatus of thesympathetic nerves. (Wien. klin. Wochens, No. 50, 1920.)

Theories on the relation of drugs to nerve systems.

The relation of drugs to particular nerve systems

is not confined to those mentioned above. Thus

curari paralyses efferent nerves leaving the spinal

cord; it paralyses first somatic motor nerves, the

phrenic being paralysed somewhat later than the

nerves supplying the skeletal muscles, it then paralyses

preganglionic fibres, and in mammals it has little or

no paralysing action on postganglionic fibres. In

each case it is the end apparatus of the nerve fibres

which is paralysed. The action of nicotine in mammals •

is also in general restricted to the connexions of the

nerves leaving the spinal cord. It first stimulates the

end apparatus of nearly all preganglionic nerves (or

the nerve cells direct) and in some animals paralyses

them, it then paralyses somatic motor nerves. Amore limited relation of a drug to nerve systems was

found by Dale (1906). Ergotoxin paralyses, or nearly

paralyses, motor contractor sympathetic nerves,

whilst leaving unaffected inhibitory sympathetic

nerves and all parasympathetic nerves.

In no case, however, is there complete correspon-

dence between the actions of a drug and the effects of

nerve stimulation. And the action of some drugs

seems to have little relation to nerve systems. Thus

^o AUTONOMIC NERVOUS SYSTEM

Dixon (1903), though he found that apocodeine, as

nicotine, paralysed all efferent nerve fibres leaving

the central nervous system, found also that it para-

lysed postganglionic fibres without relation to any

nerve system. A large dose of apocodeine paralysed

both sympathetic and parasympathetic nerves sup-

plying the iris and the heart, prevented both adrena-

line and pilocarpine from affecting the intestine, and

prevented adrenaline from causing a rise of blood

pressure, in each case leaving barium chloride with

its ordinary action. On the other hand, apocodeine

did not prevent either adrenaline or pilocarpine from

having an action on the bladder. The action thus

seems to depend rather on the nature of the tissue

than on the nervous system supplying it.

Whilst I confine the following discussion to the

problem of the cause of the correlation between

adrenaline and sympathetic nerve action on the one

hand, and that of pilocarpine and the parasympa-

•thetic nerve action on the other, I consider that the

general statements are applicable to all cases of

specific drug action.

It has been repeatedly shown that adrenaline

effects continue after degenerative section of post-

ganglionic nerve fibres (Lewandowsky, Langley,

Elliott), and it has been shown that the nerve endings

in the arteries disappear in consequence of such

section. (Eugling, Langley.) It may then be taken

as certain that adrenaline acts peripherally of the

nerve endings, that is peripherally of the terminal

nerve branches. The evidence that pilocarpine also

acts peripherally of the nerve endings rest on a

narrower basis, and at the moment depends chiefly

on the observation of Anderson that pilocarpine


causes contraction of the pupil after degenerative

section of the short cihary nerves.

Anderson and myself (J. Physiol. 31, p. 423, 1904) foundthat pilocarpine on local injection caused secretion of

sweat in the foot of the cat six weeks after section of thesciatic, though less secretion than on the intact side.

Anderson (Ibid. 33, p. 414, 1905) found the effect on thepupil mentioned above, but also found that the effect of

physostigmine disappeared a few days after the nervesection. Horsley (System of Med. by Allbutt and Rolleston

7, p. 579, 1910) observed in most cases of spinal compressionin man that subcutaneous injection of pilocarpine causedsecretion of sweat in regions supplied by nerves above thecompression, but not in the lower regions. In these cases

there is no reason to suppose that the nerve endings haddegenerated, so that the decrease of irritability wouldappear to be due to the absence of normal nerve impulses as

in the salivary glands after section of the chorda tympani.Burn (unpublished observations), after degenerative section

of the sciatic of the cat on one side, found that injection of

pilocarpine in regions other than the foot caused no secretion

in the foot. The conclusion to be drawn from the secretion

caused locally by local injection is rendered doubtful by the

fact I have mentioned above that local injection of fluid

may cause secretion in the normal animal. In recent

experiments I have found a great decrease of excitability

to pilocarpine a fortnight after section of the sciatic; the

comparative effects of fluid alone and of pilocarpine solution

I have not yet determined.

Much of the selective action of chemical substances

is prima facie accounted for by the fact that the

chemical characters of the cells of any one tissue are

on the whole more alike than are those of different

tissues, and this holds also for their ph5^sical characters.

Similarly some difference in the selective action on

the cells of any one tissue would be expected since

their characters vary, thus in glandular tissues the

chemical and physical characters of a liver cell are

not the same as those of a pancreatic gland cell. In

other words it seems inevitable that the action of

chemical substances should vary in accordance with


the degree of differentiation of the different cells, and,

so far as the different groups of cells are innervated

by different nerve systems, the action of chemical

substances would then be correlated with different

nerve systems.

But this explanation of specific drug action is in-

sufficient in itself when the same tissue is innervated

by the two nerve systems, and a drug produces the

effect of one and not of the other. A priori the result

might be due either to the nervous systems being

different and producing a definite and different change

in the cells in which they end, independently of the

nature of the cells, or to a differentiation of cell

substance independently of the nature of the nerve

systems.

So far as the action of adrenaline on unstriated

muscle is concerned the former theory has been taken

by Elliott.

Brodie and Dixon (1904), in discussing the point of

action of adrenaline and of some other drugs, had

taken the nerve and muscle to be continuous, and the

substance at the junction—the myo-neural junction

—

to be partly nervous, but under the trophic influence

of the muscle as well as of the nerve. This theory

altered to the least possible extent the older theory

that drugs acted on the terminal branches of the

nerve—a theory I had contested in the case of

nicotine. Elliott took a similar view with regard to

the action of adrenaline on unstriated muscle, and

considered that if a sympathetic nerve joined an

unstriated muscle cell it produced at the myo-neural

junction a substance responsive to adrenaline without

reference to the inherent properties of the muscle.

If in accordance with present evidence we take the


nerve impulses to be the same in sympathetic as in

parasympathetic nerves, the change produced by the

sympathetic must, on this theory, be due to its special

chemical characters.

Later (1907) Elliott considered that the action of nerveon cell could be equally well explained as the synapticmembrane theory, i.e. with discontinuity of nerve and cell.

This makes the chemical action required much less likely,

for it involves the secretion of a sujjstance from the nerveendings.

The theory, however, breaks down as a general

explanation of the relation of adrenaline to the

sympathetic and of pilocarpine to the para-

sympathetic system, for it requires that adrenahne

should cause secretion of sweat, and it does not,

and that pilocarpine should not cause secretion of

sweat or contraction of the uterus, and it causes

both.

The theory I have put forward is the latter of those

mentioned above, viz., that the effects of the drugs

are due to cell differentiation in phylogenetic develop-

ment independently of the nature of the nerve. I

take it that at an early stage, the organism consisted

of the elements of the tectal and bulbo-sacral systems

and cells connected with them, and the elements of

the epidermis. Since these developed under a com-

mon chemical and physical environment, they had

certain characters in common which enabled them

to be acted upon to a varying extent by parasympa-

tho-mimetic drugs. At a later stage the sympathetic

nerve system developed, and the cells with which

they became connected had then for the most part

acquired other chemical and physical characters

(probably in part in consequence of the presence of

adrenaline), which enabled them to be acted on by


sympatho-mimetic drugs. The effect of the sympa-

thetic nerves depended on how far the cells with

which they became connected had acquired the newer

characters, and on how far they retained the earher

ones. Thus the sweat glands of the mammal, in

consequence of the special chemical character of the

epidermic cells from which they developed, did not

become responsive to adrenaline, and retained the

early character which enabled them to respond to

pilocarpine. Since they developed late they became

connected with the sympathetic system. Amongenvironment conditions are to be reckoned nerve

stimuh. These produced a change which was at its

maximum at the point where the impulses entered a

cell and extended a variable distance beyond it, the

whole forming the neural region of the cell. The

change, which sometimes increased the responsiveness

of the cell to certain drugs, and sometimes gave rise

to it, was dependent on the nature of cell, and would

have been produced by stimuli proceeding from any

nerve. The known physical characters of drugs are

insufficient to account for the effects they produce,

though they account for a difference in rate of action

;

in consequence I consider that there is a chemical

combination between the drug and a constituent of the

cell—the receptive substance. On the theory of

chemical combination it seems necessarily to follow

that there are two broad classes of receptive sub-

stances ; those which give rise to contraction, and those

which give rise to inhibition. I assume that both

parasympathetic and sympathetic nerves can become

either contractor or inhibitor, and that the effect

produced by stimulating them depends upon the

relative amount of contractor and inhibitor receptive


substance connected with them in the cells. In

general, the sympathetic nerves became contractor

or inhibitor as the bulbo-sacral nerves were inhibitor

or contractor. The minor differences mentioned

earlier between the degi-ee and nature of the effect

produced by sympathetic nerve stimulation andadrenaline I attribute to the nerve impulses affecting

receptive substances not affected by adrenahne. As

a pure hypothesis I have suggested that the receptive

substance is—on the lines of Ehrlich's theory—-a side

chain of the molecule of the general cell substance,

and that it is in looser combination in the neural

than in the non-neural region.

On this theory it is intelligible that pilocarpine

should act not only on all tissues innervated by

parasympathetic nerves, but also on some tissues

which are innervated by the sympathetic only. Andit is not inconsistent with the possibility that adrena-

line may act on some tissues which are not innervated

by the sympathetic, or that acetyl-choline and

histamine may cause vaso-dilation by acting on a

receptive substance in capillaries receiving no efferent

nerve fibres.

The present state of knowledge does not seem to meto allow sure conclusions to be drawn as to the course

of differentiation in tissues which have a double

innervation. The chief difficulty is that it is not

known whether the cells of these tissues are of two

kinds, one innervated by the parasympathetic system

and the other by the sympathetic, or whether each

cell receives nerve fibres from both systems. The

hypothesis that the nerve fibres of the two systems end

in separate cells, is, as I have said earlier (1905), that

which is easier to bring into harmony with the general


theory given above. It is a priori probable that a

certain degree of increased complexity of chemical

constitution should cause cell division in which one

set of cells would retain the original characters and

remain connected with parasympathetic nerves which

another set would have wholly, or mainly, newer

characters and become connected with later developed

nerves—the sympathetic. But this theory involves a

conduction both of contractor and of inhibitory effect.

The simplest case is that of the bladder of frog ; con-

traction of this is caused both by sympathetic and

by parasympathetic nerves, and so far as can be seen

all the muscle fibres contract, though on sympathetic

stimulation relatively feebly. The cells do not form

two layers, so that there is no question of one layer

being innervated by one system and the other layer

by the other system. Since contraction of one part

may occur without contraction of the whole, the cells,

on the hypothesis, must be intermixed, and there

must be conduction with a rapid decrement from cell

to cell. Some evidence that there are two sets of

cells in the mucous salivary glands is afforded by the

unequal action on the cells of stimulating either the

chorda tympani or the sympathetic, and some evidence

of conduction from cell to cell is afforded by the fact

that stimulation of bulbar nerves increases the

secretion obtained by subsequent stimulation of the

sympathetic.

I suggested that the bulbar and sympathetic nerves maysupply different gland cells in the salivary glands in 1889

(J. Physiol. 10, p. 328) on the basis of the "augmented"secretion and in 1890 (J. Physiol. 11, p. 153) on the basis of

the histological changes in the glands.

There are other questions which are not yet settled.

Although perhaps we need have no serious doubt


that the primary action of sympathetic and para-

sympathetic drugs is on the cell in the region of the

nerve ending, it ought not to be forgotten that there

is no direct evidence that their action is not through-

out the cell. The theory of localised action, so far

as it is not simply a survival of the theory of an action

on nerve endings, is based partly on the assumption

that each cell receives sjonpathetic and parasympa-

thetic nerve fibres, and partly on the results obtained

in striated muscle. If each cell receives fibres from

both nerve systems and one nerve ending is inhibitory

and the other contractor, it seems inevitable that the

affected regions of the cells on electrical stimulation

must be different. But everi this deduction involves

an assumption of the way in which contraction and

inhibition is produced; it is conceivable that con-

traction is produced when the nerve fibres penetrate

the cell membrane, and inhibition when they remain

outside it, so that the two nerves cause opposite

electric charges on the cell membrane, in which case

the opposed effects of electric stimulation would not

show a localised action of the drugs. In striated

muscle there is, however, direct evidence of localised

action in the region of the nerve ending, and from the

general correspondence of the action of drugs on un-

striated muscle and glands to those on striated muscle,

it may be inferred with a high degree of probability,

that as there is localisation in one, there is also

localisation in the other.

Localised action of drugs on striated muscle. The chief

facts relating to this are briefly as follows:—It has long

been known that the middle portion of the sartorius muscle

of the frog is more excitable by induction shocks than the

nerve free ends, and that a small amount of curari lowers the

excitability in the middle portion without having any obvious


effect on that of the end portions, i.e. that the effect of a

small amount of curari is mainly, and perhaps entirely, in

the region of the nerve endings. Whether curari in larger

amount effects the electrical excitability of the nerve free

portion has not been settled. The experiments so far madetend to show that the action of curari, though much greater

in the middle region of the muscle, is not confined to it.

PoUitzer (J. Physiol. 7, p. 274, 1886) stated that a large dose

of curari (0-3 c.c. i p.c.) caused a decrease in the excitability

of the nerve-free end of the sartorius of the frog, as well as

in the part containing nerve endings, and Lapicque (C.R.

Soc. Biol. I, p. 991, 1906) that curari in proportion to its

amount caused a decrease in the direct excitability of muscleto currents of short duration. Keith Lucas (J. Physiol. 35,

p. 103, 1906) found the "optimal stimulus" of the pelvic

end of the sartorius to range from 37 to 63 in the uncurarised

muscle, and from 13 to 56 in the curarised muscle; the

experiments, however, were not made on the same muscles,

and Keith Lucas considered the excitability to be "practi-

cally" unchanged. Edstrom (Skand. A. Physiol. 41, p. 101,

1921) concluded from varied experiments on frog's musclethat curari in sufficient amount lowered the excitability of

the general muscle substance. In some of these experi-

ments it is possible that the diminution in excitability wasdue to potassium salts in the curari.

Keith Lucas, by observations on the "optimal stimuli"

of the sartorius (J. Physiol. 34, p. 372; 35, p. 103, 1906) andthe strength-duration curves for muscular contraction

(J. Physiol. 36, p. 113, 1907), showed that currents of acertain short duration caused contraction only when applied

to the nerve-containing portion of the muscle. This con-tinued to be obtained after sufficient curari to paralyse thenerve, but ceased after a somewhat larger dose of curari.

Since the excitability was local, the action of curari in

suppressing it was local. He showed that a large dose of

curari did not alter the form of the strength-duration curvesfor the pelvic end of the muscle, but, as mentioned above,his results indicated a slight decrease of excitability.

I found (Proc. Roy. Soc. B. 78, p. 186, 1906) that the tonic

contraction caused by nicotine in frog's muscle was purelylocal on local application, and that curari prevented theeffect of dilute nicotine, but not that of i p.c. nicotine. Onmicroscopic examination (J. Physiol. 36, p. 347, 1907) dilute

nicotine was seen only to cause tonic contraction whenapplied to the nerve ending region of a muscle fibre. Thecontracted part was spindle-shaped, and in favourable caseswas seven to eight times the length of the nerve ending


(Ibid. 47, p. 163, 1913). At the end of the muscle a con-centration of nicotine about 500 times as great was commonlyrequired to cause a local contraction, and this in no longtime killed the muscle. Thus, as regards the tonic con-traction, the action of nicotine and of curari is more or less

narrowly localised.

Boeke (Brain, 44, p. i, 1921) describes a " periterminal

"

network in the sarcoplasmic sole of the nerve ending in

striated muscle. He states that this network degenerates onnerve section, though somewhat later than the nerve endingitself, and suggests that it is, or contains, the receptive

substance of the neural region for the quick conductedcontraction.

Further, the results so far obtained do not show

whether the specific excitability to drugs in the neural

region is an inherited character, or whether it is pro-

duced in each generation by nerve impulses. There are

facts supporting each theory. If the cells are altered

in each generation by nerve impulses in such way that

they become excitable to drugs we should expect the

excitability to disappear in no long time when nerve

impulses cease. In fact the excitability of unstriated

muscle to adrenaline (and some other drugs) not only

remains after nerve degeneration, but in general

increases. On the other hand, Elliott found in two

cats that adrenaline and hypogastric stimulation did

not inhibit the bladder, and in these the sympathetic

nerves to the bladder were apparently absent. The

observations require confirmation, for the details

given are not wholly satisfactory, but, accepting them,

they suggest that the excitability to adrenaline is

acquired in each generation. Here again we are met

with the question whether the sympathetic nerves are

connected with a separate set of cells. If they are, it

is as easily conceivable that in the cases noticed by

Elhott the cells did not develop as that the nerve

fibres did not develop. The conception of inherited


characters localised in the region of the nerve endings

is not without difficulties, but some drugs have been

found to have their usual action in an embryonic stage,

in which it seems unlikely that nervous impulses can

be sent them. Pickering (1895) found that muscarine

and atropine acted on the mammalian heart in the

earliest stages, and I found (1905) that adrenaline in a

very early stage caused inhibition of the amnion of the

chick. But Pickering did not find that muscarine

had as early an action on the heart of the chick, and

Kouliabko (1904) only found a trifling effect to be

caused by adrenaline in one case in which he revived

the excised heart of a foetal infant by perfusion ; the

heart, however, was apparently in a pathological

state.

Classification of sympathetic and parasympathetic nerves

according to pharmacological action.

H. Meyer (191 2) has suggested that the cases in

which there is a lack of correspondence between the

effect of adrenaline and that of sympathetic nerve

stimulation, and between the effect of the pilocarpine

group of drugs and parasympathetic nerve stimula-

tion, are due to the central sympathetic centre sendmgsympathetic fibres to parasympathetic nerves, and to

the central parasympathetic centre sending parasym-

pathetic fibres to sympathetic nerves. On this

hypothesis the nerve fibres passing by way of the

sympathetic chain to the uterus are partly, and those

passing to the sweat glands wholly, parasympathetic

fibres. This hypothesis obviously introduces great

confusion into nomenclature, and there is no evidence

for it beyond that which may be considered to be

afforded by pharmacological action. The theory I


have put forward that the similarities and dissimi-

larities ill the action of drugs depends upon variations

in the cells occurring in different periods of develop-

ment is, I think, a more satisfactory explanation of the

facts. But, however, this may be, it is necessary for

descriptive purposes to keep to a nomenclature based

on anatomy, and if it is held that there are in parts of

the autonomic system two kinds of nerve fibres

characterised by having a different end apparatus, this

can be expressed by the understandable—though

inelegant—words, cholinophil and adrenophil.

Pharmacological action has led to another extension

of the term parasympathetic, in connexion with the

tone of skeletal muscles, the dependence of which on

the sympathetic I discuss presently. E. Frank

(1920) started from the fact that atropine abolishes

certain muscular tremors and tone in nervous diseases

in man, and from the generally accepted statement

that atropine stops the fascicular muscular contrac-

tion caused by physostigmine. He argued that since

physostigmine and atropine stimulate unstriated

muscle and glands innervated by parasympathetic

nerves, and this effect is abolished by atropine, the

similar effects of the drugs on the striated muscles

must be due to their receiving parasympathetic nerve

fibres. It was, however, shown by Kato and myself

(19 1 5) that whilst atropine stops physostigmine

contractions which are of central origin it does not

stop those which are of peripheral origin. The dis-

tinction of sympathetic and parasympathetic drugs

lies in their peripheral effects, so that, on the argument

the absence of effect of atropine would show the

absence of nerve fibres corresponding to parasympa-

thetic nerve fibres.


Frank also concluded on various grounds that his

" pasyarmpathetic " fibres were afferent fibres con-

ducting impulses antidromically, and that the tonic

contraction they produced was in sarcoplasm. As I

have said, I regard antidromic action as a function

of certain afferent somatic fibres. Antidromic im-

pulses can be set up in all nerve fibres, whether the

impulses have any effect or not depends on the nature

of the connexion of the fibres, and on that of the cells

with which they are connected.

Schaffer (A. ges. Physiol., p. 42, 1920) supported Frank's

theory on the ground that he found Tiegel's contracture in

man to be increased by physostigmine and by pilocarpine,

a result which might be due to an action on the central

nervous system.


References to papers quoted in the text or in the following

historical account.

Oliver and Schafer. J. Physiol. 18, p. 230, 1895.

Pickering. Ibid. p. 470.

Lewandowsky. Cntrlb. Physiol. 12, p. 599, 1898. A.

(Anat. u.) Physiol., p. 360, 1899.

Boiuttau. A. ges. Physiol. 78, p. 97, 1899.

Lewando\vsky. Cntrlb. Physiol. 14, p. 433, 1900.

Langley. J. Physiol. 27, p. 237, 1901.

Dixon. Ibid. 30, p. 97, 1903.

Brodie and Dixon. Ibid. 30, p. 491, 1904.

Konliabko. A. di Fisiol. 2, p. 137, 1904.

Elliott. J. Physiol. 32, p. 401, 1905.

Langley. Ibid. 33, p. 374, 1905; Proc. Roy. Soc. B. 78,

p. 170, 3906.

Dale. J. Physiol. 34, p. 163, 1906.

Elliott. Ibid. 35, p. 367, 1907.

Frohlich and Loewi. A. exp. Path. 59, p. 34, 1908.

E. Frank and Isaac. Zts. exp. Path. u. Ther. 7, 1909.

H. Meyer. Deut. Zts. f. Nervenhk. 45, 1912.

Lewandowsky. Zts. ges. Neurol. 14, p. 281, 1913.


Kuno. J. Physiol. 49, p. 139, 1915.Langley and Kato. Ibid. p. 410, 1915.Dieden. Zts. Biol. 66, p. 387, 1916.

Dale and Richards. J. Physiol. 52, p. no, 1918.

E. Frank. Berl. klin. Wochens, 1920. No. 31 Verh. Gesells.

Deut. Nervenartze. 10, p. 146, 1921.

Langley. J. Physiol. 55, 1921.

Luckhardt and Carlson. Amer. J. Physiol. 56, p. 72, 1921.

Degeneration of nerve endings.

Eugling (following up an observation of P. Hoffmann). A. ges.

Physiol. 121, p. 275, 1908.


Historical. The early observations on Addison's disease

and the results of extirpation of the suprarenal bodies led to

the opinion that these bodies had a special relation to the

activity of skeletal muscle. Oliver and Schafer, in their

pioneer observations on the effect of extracts of the suprarenal,

found that it caused contraction of small arteries, increased

force of the heart beat, and had a veratrine-like action onstriated muscle. These results directed attention to the

action of the extracts on unstriated muscle in general.

Lewandowsky (1898) found that the extract caused contrac-

tion of the unstriated muscle of the eye like that produced bystimulation of the sympathetic. He obtained (1889) no effect

on the intestine or bladder. He made experiments to deter-

mine whether the action was direct on the muscle or not, a

question on which there was difference of opinion. Hedecided for direct action, for he found that the effect of the

extract persisted for some weeks after excision of the superior

cervical ganglion. Boruttau (1889) noted inhibition of the

intestine. Lewandowsky (1900) observed inhibition of the

bladder. The action of adrenaline so far described was almost

identical with that of nicotine, and the references which were

made to the sympathetic were, I think, either descriptive or

made as showing obviously that the effect was on unstriated

muscle and not on striated muscle. Lewandowsky, to judge

from a late account (1913), used "sympathetic" to include the

cranial and sacral autonomic nerves. I investigated the

action of suprarenal extract (1901), and found that it caused

secretion from salivary and certain other glands. From its

effects on a large number of tissues I concluded that it did not

in any case produce the effects produced by stimulating


cranial or sacral autonomic nerves, and that it did produce in

nearly all cases the effects produced by sympathetic nerves.

The sympathetic effects not produced by the extract were

contraction of the pupil except with a larger dose, absence of

preliminary contraction of the bladder of the cat, and the

absence of secretion of sweat. The degree of effect on arteries

I found to be, in general, proportional to the degree of sympa-

thetic nerve effect. I confirmed and extended Lewandowsky's

observation of the continuance of the effect of the extract after

degeneration of postganglionic sympathetic nerves.

The action of adrenaline was to me a special case of the

problem of the specific excitability of nerve endings, a theory

at the time practically universally adopted. I had argued a

little earlier (1901) that this theory was not true as regards

the action of nicotine upon nerve cells. Thus evidence that

suprarenal extract—the effects of which were so naturally

explained by an action on nerve endings—did not act on nerve

endings was important evidence against the general theory.

Until this question was settled the exact point of action was a

secondary matter. I had, however, pointed out that the

stimulating and paralysing action of nicotine was not neces-

sarily on the whole cell substance ; and it was obvious from the

experiments I had given on the effect of adrenaline that it mustact on some special constituent of the peripheral cell. Thequestion of the point of action was taken up by Dixon (1903).

He found that apocodeine paralysed vaso-motor nerves, andthat adrenaline had then no effect on blood vessels, thoughbarium chloride still caused contraction. He concluded that

adrenaline acted on the nerve endings. In the following year

Brodie and Dixon, in the course of an account of the vaso-

motor nerves of the lungs, discussed the point of action of

adrenaline and of various other drugs. They stated that the

action of adrenaline was invariably that which was producedby sympathetic excitation. They upheld the conclusion

arrived at by Dixon that it acted on the nerve endings, andsuggested that in the experiments of Lewandowsky andmyself sufficient time had not been allowed for degeneration.

Brodie asked me for criticism of these results. I told him that

pilocarpine caused secretion of sweat six weeks after section

of the sciatic, and said there was probably something at the

nerve endings which did not degenerate when the nerveendings degenerated. Brodie, thereupon, inserted a footnote

in the paper, in which he abandoned the view that the action


was on the visible nerve ending, and, adopting the theory of

continuity, considered that the point of action was at the

junction of the nerve and muscle—the myo-neural junction—and that this was in trophic connexion both with nerve andmuscle. The myo-neural junction theory was only partially

supported by Dixon, who continued to advocate the not

improbable theory that some of the drugs mentioned in the

paper acted on the anatomically visible nerve endings.

Elliott (1905) gave further instances of the correspondence

between adrenaUne action and sympathetic action. Heshowed that the effect of the S57mpathetic on the bladder varied

in different animals, and that there was a similar variation in

the action of adrenaline. He found that contraction of the

pupil of the dog, which a small dose of adrenaline causes, wasnot obtained after destruction of the mid-brain, and he showedthat the action of adrenahne continued for long periods after

nerve degeneration. He regarded an effect or absence of

effect of adrenaline as a certain test of the existence or non-

existence of sympathetic innervation in unstriated muscle.

His general theory, which was an elaboration of that given

by Brodie and Dixon, I have mentioned in the text. In the

same year it was shown by Anderson that pilocarpine caused

constriction of the pupil after degeneration of the short ciliary

nerves. In this year also, and in the following, in an account

of the point of action of curari and nicotine on striated muscle,

I put forward the theory of the existence of receptive sub-

stances, and their formation at different times in phylogenetic

development.

Since that time further instances of the correspondence of

the action of adrenaline and of sympathetic stimulation have

been found in different vertebrates, and also instances, most of

which are mentioned in the text, in which the correspondence

is more or less uncertain.

The theory of the specific relation of pilocarpine and of

atropine to the parasympathetic system was put forward byFrohlich and Loewi (1904). Other bodies were gradually

classed with these as parasympathetic drugs. E. Frank and

Isaac (1909) and H. Meyer (1912) advocated the view that

choline had a function with regard to the parasympathetic

system similar to that which adrenaline is commonly held

to have with regard to the sympathetic. The variations in the

use of "parasympathetic" have been mentioned in the text.

Dale and Richards (1918) came to the conclusion that the


vaso-dilator action of adrenaline on capillaries was not corre-

lated with any nerve system.

I have discussed the action of pilocarpine since that is the

most familiar "parasympathetic" drug, but the action of

choline, since it is formed in the body is the most important.

The fall of blood pressure caused by choline was described byMott and Halliburton (Phil. Trans. London, 191 B., p. 211,

1899). For the effects o^ acetyl-choline cp. Dale (J. Pharm.ex. Ther. 6, p. 147, 1914) and Reid Hunt (Amer. J. Physiol. 45,

pp. 163, 231, 1918). Both consider the vaso-dilator action of

acetyl-choline to be independent of any nerve system.

It is outside the scope of my purpose to discuss how far

adrenaline and choline act normally in the body, though somereference to the action of choline on the intestine will have to

be made later. But it is clear that if excessive production of

either occurs it will cause a whole group of symptoms similar

in the main to those caused by sympathetic or by para-

sympathetic nerves. Eppinger and Hess (Die Vagotonia,

Berlin, 1910) state that pathological cases occur of hyper-

excitability (and of hyperactivity) of the whole sympathetic

system (sympathotonia), or of the whole parasympathetic

system (paras3rmpathotonia, vagotonia), and that the hyper-

excitability can be detected by the increased effect of adrena-

line and pilocarpine respectively. Their view that increased

excitability of one is accompanied by decreased excitability

of the other has not found much support.

In fish, chromaffine cells are relatively more developed than

sympathetic nerve cells, and, in consequence, the theory that

tissues are affected by adrenaline which are not affected bysympathetic stimulation has found some favour as concerns

fish.

5. The Tissues Innervated.

All cells, except those which are specially differenti-

ated for receiving afferent impressions, and those

which migrate, may be regarded as potentially able

in phylogenetic development of becoming connected

with efferent nerve fibres, the final state depending

upon the use to the organism. Experiment alone can

determine whether such connexion has developed, or,

if once developed, whether it has been retained.

Cardiac muscle, nearly all unstriated muscle, most

glands and some pigment cells have been shown to be

affected by efferent autonomic nerves. Whethercapillaries are influenced by efferent nerves is a

question for discussion. But it is obvious that

autonomic nerves have a different degree of effect on

the different cells on which they act. The different

degree of effect is very great in glandular tissue ; thus

nerve stimulation has a prompt and great effect on

the salivary glands and little or none on the intestinal

glands. And the anterior lobe of the pituitary gland

which arises from the primitive pharynx has not been

shown to react to nerve impulses. Unstriated muscle

in most cases responds promptly and greatly to nerve

stimulation either by contraction or by inhibition,

but veins do not contract as readily as arteries, nor

large arteries as readily as arterioles. The difference

in response is partly due to the relative amount of

muscle, partly to the relative degree of innervation,

but it is in part due either to the muscular excitability

being different, or to a difference in the ease with

57


which nervous impulses pass to the tissue. A certain

degree of lack of excitability, or of increase of nerve

block, would cause the tissue to be uninfluenced by-

nerve impulses; it is, for example, still doubtful

whether the small cerebral arteries are affected by the

—apparently efferent—-nerve fibres accompanying

them. The question of nerve connexion with one

element of a tissue and not with others arises more

strikingly in the case of the black pigment cells

—

melanophores. The melanophores are usually con-

sidered to be connective tissue cells. There is good

evidence that the melanophores of the skin in lower

vertebrates are influenced by sympathetic nerve

fibres, whilst the deeper lying melanophores and un-

doubted connective tissue cells are not known to be

influenced. Apparently then nervous connexion has

developed in one type only of a large class of cells.

Spaeth (19 1 6), however, chiefly as the result of obser-

vations made on the melanophores of the skin of a

fish, considers that the skin melanophores are modified

unstriated muscle cells. He lays stress on the fact

that the clumping and dispersion of the pigment

granules is brought about by reagents which cause

contraction and relaxation of unstriated muscle, and

on apparent transition forms between branched

melanophores and the unstriated muscle of the iris.

The difficulty of this view is that the histological

resemblance between cutaneous melanophores and

deep lying melanophores is greater than the resembl-

ance between cutaneous melanophores and the un-

striated muscle of the iris. And many observers

consider that the movement of granules is due to

internal movements of the protoplasm, and not as in

muscle to movement in bulk. Further, if, as has been

THE TISSUES INNERVATED 59

said by Lieben, adrenaline causes a slight movementof the pigment in the deep lying melanophores, the

much greater response of the cutaneous melanophores

might be due to the development by nerve impulses

pf this capacity to respond.

The details of innervation of the tissues will for the

most part be considered under the separate nervous

systems, but the question whether the cutaneous

pigment cells, capillaries and striated muscle are

innervated by any part of the autonomic nerve system

will be most conveniently taken here.

Pigment cells.

Variations in the tint of the skin of lower verte-

brates are brought about chiefly by the shifting of the

position of melanin granules in the melanophores.

When the skin is dark the melanophores are visible as

branched cells, having melanin granules in their

processes as well as in their cell bodies ; when the skin

is light the melanophores are visible as roundish black

clumps. In this section we are only concerned with

the evidence that nerve impulses affect directly the

melanophores, so that it is unnecessary to discuss

whether the concentration of the melanin granules is

brought about by retraction of the cell processes or by

a passage of the granules from the processes to the

body of the cell without retraction of the processes, or

whether movement of granules first occurs and

retraction of processes later, all of which views are

held. In the following account I speak of the con-

centration as contraction without prejudice to the

question whether the cell processes are withdrawn

into the body of the cell or no.

6o AUTONOMIC NERVOUS SYSTEM

That nervous impulses either directly or indirectly

cause contraction of melanophores in many lower

vertebrates is shown by the facts that in appropriate

conditions afferent impulses from one part of the body

cause the skin of the rest of the body to become lighter

in tint, that section of a cutaneous nerve causes the

skin to become darker, and that stimulation of the

peripheral end of the nerve causes the skin to become

lighter. The promptitude and the constancy with

which an effect is produced by the different forms of

experiment vary in different animals and in different

species of the same animal. They are also influenced

by local conditions which act directly on the melano-

phores, and which tend to cause them to contract or

expand, such as variations in moisture, pressure,

light, the presence of chemical bodies, and probably

temperature. In general, nerve section appears to

have a prompter and greater effect on fish than on

the frog, whether this is due to the normal tone being

greater in fish is uncertain.

The contraction of melanophores caused by nervous

action has been taken by nearly all observers to be due

to a direct action on them, but some have attributed

it to an indirect effect of change in blood supply. It

was shown by Lister (1858) on severance of all the

tissues of the thigh of a frog except the sciatic nerve

—

the artery being cut between two ligatures—that the

skin became pale. Biedermann ( 1 892) noted that the

expansion of the melanophores on ligaturing the

artery might begin in a few minutes, and become

maximal in J-J an hour, and that on ligature of the

vein the effect was produced more slowly. Both

observers, however, considered that the nerves had a

direct action. This view is based rather on a balance

THE TISSUES INNERVATED 6i

of probability than on conclusive experiment. Theconclusive experiment would be to show that nerve

stimulation causes contraction of the melanophores

when there is no change in blood supply, and when no

metabolic products which can cause contraction are

formed. The experiment which most nearly fulfils

these conditions in the frog is that made by Konigs

(191 5). Konigs states that by stimulating the sciatic

with currents of a certain voltage, duration and

rhythm, there is contraction of melanophores without

alteration in blood flow, but as the frogs were not

curarised the possibility'' of an action of muscular

metabolic products on the melanophores is not

definitely excluded, though it is true that the experi-

ment of Dutartre mentioned below and other similai

experiments makes this unlikely. In several accounts

it is implied that nerve stimulation in the frog causes

darkening of the skin more quickly than anaemia, but

accurate data are wanting. In fish the effects of

anaemia have been less studied, but so far as the

observations go they suggest a separation of pigmento-

motor and vaso-motor effect. Pouchet (1875) found

that section of the inferior maxillary nerve in the

turbot caused contraction of melanophores and that

ligature of the inferior maxillary artery did not.

Barbour and Spaeth ( 1 9 1 7) describe the melanophores

of Fundulus heteroclitus removed from the body as

long remaining expanded in Ringer's fluid. On the

other hand v. Frisch (191 1) states that anaemia in

fish causes great pallor, and it is known that the

melanophores contract after death.

The pigmento-motor nerve fibres in all probability

belong entirely to the sympathetic system. Pouchet

(1876) found in the turbot that section of the ventral


division of a spinal nerve distal of its ramus com-

municans caused darkening of the skin in the area

supplied by the division, and that section centrally

of the ramus did not. The result was confirmed by

Rynberk (1906) in the sole, and by v. Frisch (191 1) in

the minnow. Dutartre (1890) found in the frog that

stimulation of the upper part of the spinal cord after

section of all the rami communicantes on one side

did not cause on this side change of tint of the skin.

Possible antidromic action has not been looked for.

The theory of direct action is to some extent sup-

ported by the microscopical observations of Ballowitz

(1893) and others, who describe nerve endings on the

melanophores of fish; but nerve endings are also

described on the cells of the choroid plexus, on which

nerve action is unknown, and on capillaries on which

it is doubtful. Lieben (1906) found that adrenaline

applied to the surface of the frog's web caused con-

traction of the melanophores without change of blood

flow, and Barbour and Spaeth (191 7) that it caused

contraction of the melanophores of Fundulus after

removal from the body.

From the results mentioned above, taken as a whole,

it may be fairly concluded that nerve fibres cause

contraction of melanophores by direct action.

Carnot (1896) put forward the hypothesis that the

melanophores are supplied with inhibitory as well as

with contractor fibres. The basis for this was

apparently that some substances, e.g. amyl nitrite

caused expansion of the melanophores of the frog.

Whilst this fact is no evidence for the hypothesis, there

are other facts which are at least suggestive. Pouchet

described the melanophores in the turbot after nerve

section as not being completely expanded, and


Barbour and Spaeth observed that adrenaHne appUedafter ergotoxine to the melanophores of excised scales

of Fundulus caused expansion instead of contraction.

It may be noted that they found sUght expansion

to be caused by pilocarpine.


Lister. Phil. Trans. London, 148, p. 607, 1858.

Pouchet. J. de I'Anat. Physiol., p. 113, 1876.

Dutartre. C. R. Acad. Sci. in, p. 610, 1890.

Biedermann. A. ges. Physiol. 151, p. 455, 1892.

Ballowitz. Zts. wiss. Zool. 56, p. 673, 1893 (nerve endings

in Teleostean melanophores).

Carnot. C. R. Soc. Biol., p. 927, 1896.

Lieben. Zntrlb. Physiol. 20, p. 108, 1906.

Rynberk. Ergeb. Physiol. 5, p. 429, 1906.

V. Frisch. A. ges. Physiol. 133, p. 319, 1911.

Konigs. Etude de I'excit. d. nerfs, etc., Paris, 1915.

Spaeth. J. Exp. Zool. 20, p. 293, 1916.

Barbour and Spaeth. J. Pharm. exp. Ther. 9, p. 356,

1917.

A full summary of the numerous papers on pigment cells up

to 1906 is given by Rjmberk, see also Konigs. For papers

on fish melanophores, 1906 to 1911, see v. Frisch, and for later

papers Ballowitz (A. ges. Physiol. 157, p. 165, 1914). Manyof the observations deal with the adaptation of tint to surround

ings. Adaptation has a less direct bearing on the question

of the existence of pigmento-motor fibres than nerve section

and stimulation, since it involves controversy as to the degree

of direct action of light on the melanophores ; in consequence I

have not thought it necessary to discuss it.

The red chromatophores which occur in some animals are

influenced by nerve stimulation in the same way as the black

chromatophores. It is doubtful whether there is any nerve

effect on yellow chromatophores or on guanin cells; concentra-

tion of particles in both cells when exposed to light has been

described by Ballowitz.


Sympathetic nerves and the innervation of capillaries.

There is no doubt that constriction and widening of

capillaries can be caused by chemical substances

independent of variations in the blood pressure, and

there is little doubt that this may be produced by an

action on the epithelioid cells forming their walls. Since

nerve fibres, in part of sympathetic origin, run from

the small arteries and accompany the capillaries, it is a

natural hypothesis that the sympathetic controls the

size of the capillaries in the same way that it controls

the size of the small arteries. The evidence of such

action is, however, very slight. The only instance in

which capillary contraction, certainly independent of

pressure variation, has been obtained on nerve

stimulation is that found in the nictitating membrane

of the frog by Steinach and Kahn (1903). But the

effect they obtained was a longitudinal folding of the

epithelioid wall, and they attributed it to contraction

of cells outside the capillaries, probably they con-

sidered the cells described by Rouget (1873) and by

S. Mayer (1886), and not to a contraction of the

epithelioid wall. Rouget and Mayer described

branched cells, which they held to be muscle cells

surrounding certain capillaries. It is possible that

some capillaries, especially in a certain stage of

development, have scattered muscle cells in their

outer coat, but there is no evidence that this is

of more than a local occurrence in amphibia, and

none that it occurs in fully formed mammaliancapillaries. In any case, Steinach and Kahn's ob-

servations afford no support to the theory that the

sympathetic send efferent fibres to the epithelioid

wall of capillaries.

THE TISSUES INNERVATED. 65

Many observers have watched the capillary circula-

tion in the web of the frog during sciatic stimulation,

and no one has found that the stimulation causes anycertain contraction in them. The difference in the

eifect on the arteries of the web and that on the

capillaries and veins when the sympathetic nervefibres supplying it are stimulated is most striking.

The arteries contract and stop the blood flow com-pletely or nearly completely, whilst the capillaries

remain fairly full; sometimes, however, the capil-

laries become empty of blood corpuscles and their

diameter decreases. This was noted by Lister (1858)

on stimulating the spinal cord. He attributed the

effect to the arterial contraction being sufficient to

stop the blood corpuscles, but insufficient to stop the

plasma. I found (1911) on prolonged (and repeated)

stimulation of the sympathetic, or of the sciatic nerve

after curari had been given, that sometimes the capil-

laries gradually became empty instead of remaining

partly full. These variations in capillary diameter

are, I think, more probably caused by variations in

the balance of pressure inside and outside the capil-

laries and by the action of metabolites than by direct

nerve action. The capillaries are surrounded bylymph spaces, and increased pressure in these would

tend to compress the capillaries. The conclusion

arrived at by Roy and Graham Brown (1879), that

capillary diameter is little influenced b}'' variations in

arterial pressure, is no doubt true for the circulation

in the web of the frog in ordinary conditions. The

long-known fact that in the web great contraction of

the arteries is not accompanied by any considerable

diminution in the diameter of the capillaries shows

that the normal pressure distends them but little, and


that the extra-vascular pressure is slight. But I

do not agree with the statement of Krogh (1920) that

arterial pressure has no appreciable effect on capillary

diameter; it has seemed to me that there is a distinct,

though slight, effect both when pressure falls owing to

local contraction of arterioles, and when it rises,

owing to contraction elsewhere. And I take it that

there is at times a local increase in extra-capillary

pressure capable of considerably compressing the

capillaries. However, this may be, it is certain that

if any contraction of capillaries is produced by nerve

stimulation in the web of the frog it is of quite a

different order from that which the nerves produce on

the arterioles.

In the mammal the facts are different. It has long

been noticed that stimulation of the cervical sympa-

thetic usually causes the ear, with the exception of

the veins, to become completely pale, so that the

capillaries as well as the arteries are emptied of blood.

This has generally been taken as showing that whenthe arterial contraction cuts off the blood supply, the

pressure in the tissue spaces has been sufficient to

empty the capillaries. Rouget (1879), however, con-

sidered that the pallor of the face caused in man by

emotions could only be due to capillary contraction.

In recent years there has been a revival of inquiry

into the conditions of capillary circulation. Dale and

Richards have shown that histamine causes intense

dilatation of the capillaries. Cannon pointed out

that in shock there was stagnation of blood in the

capillaries. Krogh found local dilatation on mechani-

cal stimulation of the tongue of the frog, and the old

observation of the red stripe caused by stroking the

skin has received renewed investigation. These


evidences of independent action of the capillaries,

though all in the direction of vaso-dilation, have

revived attention to the question of vaso-constrictor

nerves for them, and Hooker (1920) interprets the

pallor of ear (cat) caused by sympathetic stimulation

as involving active capillary contraction. No satis-

factory evidence is, however, adduced to show that

this constriction is not passive. It will be noticed

that if the pallor in the tissues of the mammal is caused

in part by contraction of the capillaries, the action of

nerves on them must be nearly as prompt as that on

the arterioles.

The question of the production of capillary dilata-

tion by afferent nerve fibres will be dealt with in the

section on antidromic action. The action of adrena-

line has been referred to above (p. 32).


Lister. Phil. Trans. Roy. Soc, Pt. 2, p. 607, 1858.

Rouget. A. Physiol, norm, et path 5, p. 656, 1873.

Roy and Graham Brown. J. Physiol. 2, p. 338, 1879.

Rouget. C. R. Acad. Sci. 88, p. 916, 1879.

S. Mayer. Wien. Sitzb. 93. Abt. 3, p. 45, 1886; Anat. Anz. 21

p. 442, 1902.

Steinach and Kahn. A. ges. Physiol. 97, p. 105, 1903.


Hooker. Amer. J. Physiol. 44, p. 30, 1920.

Krogh. J. Physiol. 53, p. 405, 1920.

Capillary contractility. The theory that the capillaries

receive vaso-constrictor nerves from the sympathetic takes for

granted that the capillaries are contractile. The direct

evidence of capillary contractiUty, except in response to

chemical substances, is limited to a few tissues chieflyj_in

amphibia. The discussion of the question is of long standing.

Allen Thomson in 1835 (Todd's Cyclopaedia Anat. and Physiol.

I, p. 671) discussed the evidence and supported the theory.

In the subsequent 30 years most observers inclined to the

opinion that the variations known to occur in the diameter of


the capillaries were due either to variations in blood pressure

or to variations in the elasticity of the wall caused by in-

flammatory processes and by reagents. A change of opinion

began with Strieker's observations in 1865 (Wien. Sitzb. 51,

Abt. 2, p. 16). He found spontaneous variations in diameter

in the capillaries of the nictitating membrane of the frog and(ibid. 52, p. 379, 1866) contraction in them on stimulating with

strong induction shocks. Similar stimulation also caused

contraction in the capillaries of the tail of young tadpoles, but

there was usually no effect in fully grown tadpoles, and very

rarely in the capillaries of the mesentery of the frog. Golubev

(A. mik. Anat. 5, p. 49, 1869) andTarchanov (A. ges. Physiol. 9

p. 407, 1874) confirmed the constriction of the lumen of the

capillaries of the nictitating membrane of the frog on electrical

stimulation, but attributed it to a swelling of spindle-shaped

elements, later shown to be nuclei. Strieker (Wien. Sitzb. 74,

Abt. 3, p. 313, 1876) confirmed his previous results, and added

that strong induction shocks caused contraction of capillaries

in fully grown tadpoles after they had been treated with dilute

alcohol. He observed sometimes the bulging of the spindle-

shaped elements, as described by Golubev, but noted contrac-

tion of the whole tube in young tadpoles. Severini's papers

(1878-9) I have not been able to obtain; he is quoted bynumerous writers as having shown that bulging of the nuclei,

and often narrowing of the lumen of the whole capUlary is

caused by oxygen and the reverse change by carbonic acid.

Roy and Graham Brown (1879) considered that the diameter

of the capillaries was regulated by metabolites probably acting

directly on the capUlary wall. The opinion that the capil-

laries were contractile had been supported by Rouget, but the

contractile part he held were branched muscle cells surrounding

the capillary tube. Such cells he described as present in the

hyaloid membrane of the frog (A. Physiol, norm, et path.,

p. 601, 1873). Later (1879) he found similar cells around the

capillaries of the tail of the tadpole, the capsulo-pupiUary

membrane of new-born mammals, and of the omentum of

young mammals. AU these, he said, contracted on electrical

stimulation. S. Mayer (1886, 1902) confirmed Rouget's

observations, which had received little attention, and des-

cribed branched cells as being present around the capillaries

of the intestine and of the bladder of the salamander and frog.

Some at any rate of these cells were almost certainly connec-

tive tissue cells. Steinach andKahn (1903) , on direct stimulation


of the capillaries of the perioesophageal membrane of thefrog and of the omentum of the kitten, did not obtain con-traction in capillaries of less than about 10 ju in diameter.

Krogh (J. Physiol. 52, p. 457, 1919) has given reason to

believe that in resting striated muscle, some of the capillaries

are too small to admit blood corpuscles, but that they dilate

during contraction and on massage. In general, he holds that

the capillaries are constantly undergoing active changes in

diameter.

In mammals ocular evidence of contractility is given bymechanical stimulation (see Ebbecke. A. ges. Physiol. i6g,

p. I, 1917).

Sympathetic nerves and the innervation of

striated muscle.

In recent years there has been much discussion on

the question of the innervation of striated muscle by

sympathetic nerve fibres. Bremer, in 1882, found

that occasionally in limb muscles and frequently in

the muscles of the tongue, nerve endings were formed

by non-medullated as well as by meduUated nerve

fibres. Some of the non-medullated fibres were

fibres which had lost their medulla in their course in

the muscle, some were branches of medullated fibres,

but some arose from a non-medullated plexus, and

could not be traced further centrally. The nerve

endings of the two kinds of fibres differed in appear-

ance; they were commonly, but not always, close

together, in the muscle fibre. Perroncito^ (1902) put

forward the theory that the muscle fibres were

supplied with nerve endings both from the sympa-

thetic and from the cerebro-spinal systems. Mosso

(1904) chiefly on the basis of Perroncito's observa-

tions, suggested that the sympathetic governed the tone

and slow contraction of muscle, and the cerebro-spinal

"• Quoted from Mosso. Perroncito's paper, in Arcli. Ital. de'Biol. 38,

P- 393. 1902, does not give this theory.

^o AUTONOMIC NERVOUS SYSTEM

(i.e. somatic) nerves governed quick contraction;

both forms of contraction he attributed to the

muscle fibrils. The question of the connexion of the

sympathetic with muscle fibres was brought to the

front by Boeke in a series of observations from 1909

onwards. Boeke (1912, 1917), and subsequently

Agdur, and Boeke with Dusser de Barenne (191 9),

showed by the degeneration method that a plexus of

sympathetic nerves was in fact present in muscle, and

that from this plexus nerve fibres proceeded to the

muscle fibres forming endings such as had previously

been described in normal muscle.

In these observations the somatic nerves of the eye,

finger, and intercostal muscles were cut peripherally of their

trophic centres, and after the nerves had degenerated themuscles were treated by the silver impregnation method.The results obtained by Boeke (1912, 1917) on the eyemuscles are not easy to interpret. He found that theendings of the great majority of the non-meduUated fibres

were not degenerated a few days after intercranial section of

the motor nerves, but were degenerated in about threeweeks. He considered that these fibres were autonomicfibres issuing in the nerve roots. If this is so, they aredifferent from other autonomic fibres, either in having noperipheral nerve cells on their course, or in degeneratingperipherally of the nerve cells. In view of this difficulty

rigid proof is required that the non-medullated nerves in

question do not arise from the medullated fibres, and becomeimpregnated with silver in consequence of the persistenceof the sheath (cp. p. 26). Against this possibihty it may beurged the Boeke describes the endings as being hypolemmal,and that hypolemmal fibres are not known to have a sheath.The evidence of the existence of sympathetic nerve endingsin the eye muscles was not very decisive ; it depended on thepresence of a few non-medullated fibres after degenerationof the somatic nerves, and on an apparent slight decrease ofnon-medullated fibres after degeneration of the sympatheticnerves.

It may be mentioned that Botezat (Zts. wiss. Zool. 84,1906; Anat. Anz. 35, 1910) described the muscles of birdsas having nerve endings both from medullated and non-medullated nerves; he did not, however, trace the centralorigin of the latter.


Since the sympathetic supphes the vessels of the

muscles, it is obvious that a plexus of sympathetic fibres

will remain after degeneration of the somatic nerves.

The essential point is, whether fibres from this plexus

form nerve endings under the sarcolemma of the

muscle fibres, or whether they all end in connexion

with blood vessels. Boeke, Agdur, and Dusser de

Barenne describe hypolemmal nerve endings, though

Agdur takes some to be epilemmal. But nerve

endings, definitely sympathetic, have so far been only

shown by the silver impregnation method, and if all

muscles have them it is remarkable that they have

not been described in methylene blue preparations

of the sartorius of the frog in which a certain number

of nerve endings of somatic nerves and of nerve fibres

around blood vessels are brought out with ease and

certainty. Whilst this apparent contradiction has

still to be cleared up, the evidence at present is

decidedly in favour of Boeke's views.

The histological results led de Boer to investigate

the function of the sympathetic nerves supplying the

muscles. He found (19 13) that section of the rami

communicantes of the nerves of the leg of the frog

caused loss of tone of the muscles, and he concluded

that the function of the sympathetic was to keep up

muscular tone. Pekelharing and Hoogenhuyze had

brought forward much evidence to show that muscular

tone involved an increase of creatine, i.e. protein

breakdown. The tone they considered, was brought

about by somatic nerves, and on the lines of Bottazzi's

theory that it might be sarcoplasmic. De Boer,

holding the tone to be caused by sympathetic nerves,

suggested that the sympathetic caused protein break-

down in sarcoplasm. de Boer, a little later in the


same year, found that extirpation of the lumbar

sympathetic in the cat caused a loss of tone in the

muscles of the hind limb and of the tail. The results

obtained by subsequent observers have varied greatly,

but on the whole they show that a slight decrease of

muscular tone may be caused, both in the frog and in

the mammal, by section of the sympathetic. Obser-

vations in addition to those mentioned below are

given in the notes at the end of this section.

The theory of de Boer that the tone of the muscles is

entirely due to impulses passing by the sympathetic

is not tenable. Previous experiments on the effects

of cutting the sympathetic and of excision of the

ganglia had, I think, given ample evidence that

neither operation caused complete loss of tone in the

muscles of the head and limbs. Dusser de Barenne

(1913), Negrin y Lopez and v. Briicke (1917) and

V. Rijnberk (191 7) showed that decerebrate rigidity

is not dependent on sympathetic innervation. Jansma

(19 1 5) described in the frog greater loss of tone on

cutting the sciatic than on cutting the rami com-

municantes, and Dusser de Barenne (1916) a greater

loss of tone on cutting the posterior roots than on

cutting the rami. De Boer (1915) referred the

decerebrate rigidity occurring after section of the

sympathetic to a tetanic tone caused by the somatic

nerves, but tetanic tone, if it involves the fibrillae,

causes increased production of COg, and, according to

Roaf, the COg production in decerebrate rigidity is

not perceptibly greater than that of the curarised

muscles. Further, Lilljestrand and Magnus (19 19)

found that the tone caused by tetanus toxin, shownby H. Meyer to be due to central stimulation, con-

tinues after section of the sympathetic.


When a loss of tone has been obtained bysympathetic nerve section, its duration has not been

found to be the same by different observers. Dusser de

Barenne ( 1 9 1 6) found that the loss of tone on excision

of the lumbar sympathetic in the cat disappeared in

four to eight weeks. Negrin y Lopez and v. Briicke

(191 7), in similar experiments, found the loss of tone

(when it occurred) disappeared in a day or two.

Ducceschi (1919), who observed a slight drooping of

the ear of the rabbit after excision of the superior

cervical ganglion, noted that the position of the ear

became normal in a few weeks, but that by weighting

the two ears a slight loss of tone could be detected for

six months.

Whilst the theory that the sympathetic causes some

degree of muscular tone is satisfactory in that it pro-

vides a function for the nerve endings described by

Boeke, the experiments which have been made do

not decide definitely whether the loss of tone conse-

quent on sympathetic section is, or is not, due to

concurrent vaso-dilation. The variation in degree

and permanence of effect on tone in the mammal comes

well within the limits of the variations which have

been found in vascular effects. It is true that de Boer

(191 5) has described a decrease of tone in the gastroc-

nemius of the frog on cutting the rami communicantes

after excision of the abdominal viscera, when presum-

ably the circulation was at most sUght, but in view of

the varying results which have been obtained in the

frog, further more decisive observations are required.

A defect in the evidence that the sympathetic

causes tonic contraction in muscle is that no one has

been able to obtain contraction by stimulating it.

Even the rectus abdominis and the arm muscles of


the frog, which are so prone to tonic contraction,

show no visible change on stimulation of the rami

communicantes. If, then, the sympathetic causes

tone in striated muscle, the nervous impulses produc-

ing it are different from the nerve impulses which

cause tone in unstriated muscle. The lack of effect

of nerve stimulation is the more striking in that in

mammals tonic contraction is readily produced by

stimulating the cerebellum.

The question of the nature of the changes occurring in

tonic contraction is outside the scope of the subject discussed

in this section, but there is a possibility regarding it whichbears on the question of sympathetic innervation. It is

known that there are some forms of continuous contraction

set up by the central nervous system in which the electric

state of the tissue remains constant, and in which apparently

there is no chemical change. It is conceivable that this

form of contraction should be produced by a continuous

current from the nerve cells which gives rise to a static

charge on the membrane of the contractile substance. Onthis hypothesis, on the one hand the absence or occurrence

of tonic contraction on nerve stimulation would depend onthe character of the nerve, and on the other hand the

probability that the tonic contraction is due to the fibrillae

and not to sarcoplasm would be very great.

In view of the theory of Pekelharing that muscular

tone is accompanied not by a breakdown of carbo-

hydrates, but by a breakdown of protein with forma-

lin of creatine, and of de Boer's theory that tone is

maintained by the sympathetic, observations have

been made to determine whether cutting off impulses

coming by the sympathetic decreases the amount of

creatine in muscles. Jansma (1915) found a slight

decrease in the creatine content of the muscles of the

leg of the frog, on cutting the rami communicantes of

the sciatic plexus, but the results hardly seem to be

outside the limits of experimental error. Riesser

(191 6) paralysed' the motor nerves in rabbits with a


minimal dose of curari, cut the sciatic on one side, and

then subjected the animals to conditions tending to

stimulate the sympathetic heat centre postulated byH. Meyer. He found less creatine on the side on

which sympathetic impulses had been cut off byseverance of the sciatic, but this only occurred if the

circulation through the limb was restricted by tying

the femoral arteries. The conditions in these experi-

ments are too complex to carry conviction. On the

other hand, v. Rijnberk (191 7) induced decerebrate

rigidity in cats after excision of the lumbar sympa-

thetic on one side, and found no difference in the

creatine content of the muscles of the hind legs on the

two sides, and Kahn (19 19) found less creatine in the

arm muscles of the frog during the clasping season than

in the leg muscles.

Experiments have also been made on the effect of

the sympathetic on the oxygen use and carbonic acid

formation in muscle. The starting point of these

was the observation of Frank and Voit that curari

given to dogs in not much more than the amount

required to paralyse motor nerves did not cause

increased COg production, provided the muscles were

at rest, but did, as in Pfluger's experiments, when

given in larger quantities. The larger quantity

Frank and Voit considered paralysed the sympathetic

vaso-motor nerves. Mansfeld and Lukacs (1915)

estimated the respiratory exchange in dogs, and

found that in a lightly curarised animal, section of

the sciatic caused a decrease of respiratory exchange

if the sympathetic was intact, but did not do so if the

sympathetic connexions had been cut. On their

view the section abolished muscular tone, and so

decreased the respiratory exchange of the muscles.


Section of the sympathetic would, however, tend to

decrease the amount of blood in the viscera, and so

tend to decrease the respiratory exchange. An in-

crease in CO2 formation would tend to connect the

sympathetic with carbohydrate metabolism, but no

decrease of muscle glycogen was found by Ernst (1915)

on stimulation of the sciatic in lightly curarised frogs.

Further, Nakamura (1921) determined the oxygen use

in the muscles of the lower part of the leg and obtained

no change in it on section of either sympathetic or

sciatic in anaesthetised and in decerebrated and

lightly curarised cats. The existence of tone before

section of the nerves was not,, however, directly

tested. Stimulation of the sympathetic caused a

decrease in oxygen use ; since this was accompanied

by a great decrease in blood flow, the presumption is

that it was caused by the decrease in oxygen supply,

although an inhibitory action was not definitely

excluded. There is evidence that several forms of

muscle tone, and possibly all, involve chemical change

at the onset only. So far as this occurs little or no

change in metabolism would be expected to occur on

abolishing it.

E. Frank (1920), in connexion with his theory of the

production of tone (plastic tone) by antidromic

irapulses (cp. p. 52), considers that the sympathetic

instead of increasing tone inhibits it. The evidence

given is that adrenaline abolishes certain forms of

muscular twitching and tone in man, that it abolishes

in the dog the awkwardness and rigidity caused by

injecting physostigmine, and that Schaffer found

Tiegel's contracture to be abolished by injecting

adrenaline. Schaffer took ergographic tracings of the

contraction of the muscles of the forearm in a man


in which Tiegel's contracture was marked. Heinferred that the abolition of the contracture byadrenaline was not due to vaso-constriction since the

effect was produced in two to four minutes, and in a

plethysmyograph experiment made at another time

the decreased volume of the arm only began in 5 J

minutes ; and, further, since adrenaline had a some-

what quicker effect than cutting off the blood supply

by compression of the upper arm. The theory

assumes that the action of adrenaliae is in all cases on

the end apparatus of sympathetic nerves, and in the

particular instances that the effect produced by

adrenaline is not produced either by central action

or by peripheral vaso-constriction. In Schaffer's

experiments a greater decrease of blood supply would

be caused by adrenaline than by compression of the

upper arm. As mentioned above (p. 33), several

observers have been unable to find that adrenaline

has any effect on normal muscle.

It has been suggested that the pseudo-motor action

of the chorda tympani on the tongue described by

Vulpian and by Heidenhain indicates that para-

sympathetic fibres may form nerve endings in con-

nection with striated muscle. The pseudo-motor

effects in this and other regions is, I think, more

probably due to an action of metabolites on muscle

altered by section of its motor nerve. The facts

bearing on the question will be discussed later.

We may summarise the results as follows :—There

is a balance of histological evidence that sympathetic

nerve fibres form hypolemmal nerve endings on some

striated muscle fibres, but the evidence is insufficient

to show that they form nerve endings on all of them.

There is no satisfactory evidence that the sympathetic


directly affects the formation of creatine, or COg in

muscle. The only change in muscle which has been

found on stimulating the sympathetic is a decrease in

oxygen use, an effect which may be due to the con-

current decrease of oxygen supply. There is a balance

of evidence that section of the sympathetic nerves

supplying a muscle causes a slight loss of tone, but

conclusive evidence that this is not the result of section

of vaso-motor nerves is lacking. Until this question

is settled it is premature to discuss whether the sympa-

thetic acts on sarcoplasm or on fibrillae; it appears,

however, to be certain that no form of tonic con-

traction on striated muscle attributed to sympa-

thetic nerves is distinguishable from that produced

by somatic nerves.


Bremer. A. mik. Anat. 21, p. 165, 1882.

Perroncito. Ges. med. Ital. 1902 (quoted from Mosso).

Mosso. A. Ital. de Biol. 41, p. 183, 1904.

Boeke. Intern. Monats. Anat. Physiol. 28, p. 377, 1911.

Verhd. Anat. Gesells, 1912.

Roaf. Quar. J. exp. Physiol. 5, p. 31, 1912.

Boeke. Anat. Anz., p. 343, 1913.

de Boer. Folio Neurob. 7, pp. 358, 837, 1913.

Dusser de Barenne. Ibid. p. 651.

de Boer. Zts. f. Biol. 65, p. 239, 1915.

Jansma. Ibid., p. 365.

Mansfeld and Lukacs. A. ges. Physiol. 161, p. 467, 1915.

Ernst. Ibid., p. 483.

Dusser de Barenne. Ibid. 166, p. 145, 1916.

Riesser. A. exp. Path. u. Pharm. 80, p. 183, 1916.

Negrin y Lopez and v. Briicke. A. ges. Ph5'siol. 166,

p. 55, 1917.

Rijnberk. A. neerl. de Physiol, i, p. 727, 1917.

Boeke. Verhand. k. Akad. v. Wetens, Amsterdam. Sect. 2,1917Agdur. Proc. R. Acad. Amsterdam 21, p. 930, 1919.

Boeke and Dusser de Barenne. Ibid., p. 1227.


Ducceschi. A. di Fisiol. 17, p. 59, 1919.Kahn. A. ges. Physiol. 177, p. 294, 1919.Liljestrand and Magnus. Miinch. med. Wochens, p. 551,

1919.

E. Frank. Berl. klin. Wochens., p. 725, 1920. Deut. Zts.

Nervenheilk. 70, p. 146, 1921.

Schaffer. A. ges. Physiol. 185, p. 42, 1920.

Nakamura. J. Physiol. 55, p. 100, 1921.

Further references are given by Boeke, op. cit. 1911, de Boer,

op. cit. 1915, and Dusser de Barenne, op. cit. igi6.

Kuno (J. Physiol. 49, p. 139, 1915) did not find any loss of

tone in the muscles of the leg of the frog on cutting the ramicommunicantes running to the sciatic nerve. Salek andWeilbrecht (Zts. f. Biol. 71, p. 247, 1920), in similar experi-

ments, but in which the hind legs were suspended in cold

water, observed as a rule no loss of tone at first, but later

there was sometimes a slight loss. Cobb (Amer. J. Physiol. 4,

p. 478, 1918) found no loss of tone in the hind legs and tail of

the cat on cutting the sympathetic chain between the 4th and5th lumbar ganglia, v. Rijnberk (A. neerl. i, p. 727, 1917)extirpated the lumbar sympathetic in cats, and some time later

induced decerebrate rigidity; he found no constant difference

in tone on the two sides.

Kure, Hiramatsu, and Naito (Zntrlb. Physiol. 28, p. 130,

1914) stated that the tone of the diaphragm depended uponthe splanchnic nerves. Mansfeld (A. ges. Physiol. 161, p. 478,

1915) described a loss of tone on cutting the sciatic of the frog

after curari had been given in quantity just sufficient to

paralyse the motor nerves.

The second rise in the curve of a smgle muscular contrac-

tion and the prolongation of the contraction of a veratrinised

muscle have both been held by some observers to be sarcoplas-

mic, de Boer (op. cit. 1915) considered both to be due to

impulses passing by the sympathetic. On stimulating the

sciatic of a curarised and veratrinised muscle he obtained in

some cases a slow prolonged contraction without preliminary

quick contraction, de Boer (op. cit. 1915) also claimed that

section of the lower rami communicantes in the frog delayed

rigor mortis ; as is known, cessation of circulation in the spinal

cord leads to a general stimulation of the sympathetic,

de Boer held that sympathetic nerves running to muscle were

similarly stimulated, and by causing increased metabolism

hastened rigor. Dusser de Barenne (op. cit. igi6), however.

8o AUTONOMIC NERVOUS SYSTEM

contested the statement that section of the sympathetic

hastens rigor mortis.

Note on the action of adrenaline and of pilocarpine on the

sweat glands.

As mentioned on page 29, Dieden found that local injection

of adrenaline into the cat's foot caused secretion in certain

conditions. The conditions were nerve section or very deep

anaesthesia. The absence of secretion in other conditions he

considered was due to inhibitory nerve impulses passing

(probably) by the posterior roots. In the text, I have taken

the action of adrenaline solution as being due to the fluid in

which the adrenaline is dissolved. The reasons for this were

(i) that when o.i p.c. adrenaline caused secretion. Ringer's

fluid, locally injected, caused a greater secretion; (2) that

the secretion caused by adrenaline (when obtained) wasstrictly local, and did not occur in the foot beyond the injected

area; and (3) that i p.c. barium chloride, amyl nitrite, hyper-

tonic salt solution, or distilled water, also commonly caused a

secretion. Unlike Dieden, I did not find that nerve section

had any constant effect on the result of local injection. All

secretion, including that caused by barium chloride, wasprevented by atropine.

After the earlier sheets of this book had been printed I madefurther experiments, and in some of these, adrenaline (gener-

ally o.ooi p.c.) caused markedly more secretion than Ringer's

fluid. Whether this was due to a different depth of injection,

or to a difference in the excitability of the glands, or to an

occasional action of adrenaline remains to be determined.

The action of pilocarpine after the nerves have degenerated

does not depend on the fluid in which it is dissolved (cp. p. 41)

for the secretion is not confined to the part of the foot injected,

but spreads over the whole of the secretory surface. Thetime after denervation at which pilocarpine still causes

secretion varies greatly in different animals. This, indeed,

was shown by the early observations of Marme and of Luch-

singer (Hermann's Hdb. d. Physiol, v., Th. i, p. 428, 1883),

who occasionally, but only occasionally, obtained secretion

2-3 weeks after section of the sciatic. I may note that norm-

ally there is great variation in the excitability of the sweat

glands in different cats, occasionally pilocarpine and nerve

stimulation cause no secretion. Local injection of any fluid,

but especially of adrenaline, reduces the response of the glands

to pOocarpine.

Printed by

W. Heffer & Sons Ltd.,

Cambridge, England.

Documents

irHE AUTONOMIC NERVOUS SYSTEM - Internet Archive